1
00:00:00,960 --> 00:00:02,879
Speaker 1: Hey Richard, Hey Carl, what do you know?

2
00:00:03,120 --> 00:00:06,679
Speaker 2: Well, I know that our friend Michelle Rubusta Monte is

3
00:00:06,719 --> 00:00:08,880
with us to tell us about something that's going on

4
00:00:09,519 --> 00:00:11,199
adjacent to DEV Intersection.

5
00:00:11,240 --> 00:00:15,560
Speaker 1: What is it? It's cybersecurity Intersection. Let's let Michelle tell

6
00:00:15,599 --> 00:00:16,199
that story.

7
00:00:16,679 --> 00:00:21,320
Speaker 3: Hey Michelle, Hey Carl, Hey Richard, how are you.

8
00:00:21,000 --> 00:00:22,839
Speaker 2: Tell us about cybersecurity Intersection?

9
00:00:23,480 --> 00:00:26,920
Speaker 3: Well, so, Richard and I are partnering with the group

10
00:00:27,000 --> 00:00:30,359
that does DEV Intersection and next Gen AI, and we

11
00:00:30,600 --> 00:00:34,200
are putting on a new conference dedicated to one hundred

12
00:00:34,240 --> 00:00:40,000
percent security focused topics. And I mean, honestly, the lineup

13
00:00:40,039 --> 00:00:43,679
of speakers is incredible. We have Paula A. Jenis, who's

14
00:00:43,719 --> 00:00:47,600
here from Poland and does keynotes all over the world

15
00:00:47,719 --> 00:00:50,600
and is one of the top rated RSA speakers and

16
00:00:50,640 --> 00:00:53,159
black hat speaker. We're so lucky to have her. But

17
00:00:53,240 --> 00:00:56,000
she's not only keynoting, she's got a workshop teaches you

18
00:00:56,039 --> 00:01:01,039
about protecting your environments against hackers and shows you about

19
00:01:01,280 --> 00:01:03,719
how to you know, do attacks so that you can

20
00:01:03,759 --> 00:01:07,840
prevent them. It's pretty cool and sessions like that as well.

21
00:01:07,840 --> 00:01:10,799
But we also have speakers from Microsoft. We have we

22
00:01:10,840 --> 00:01:15,120
have speakers that specialize in you know secure coding practices,

23
00:01:15,640 --> 00:01:20,439
Azure security, zero trust architectures on Azure uh and people

24
00:01:20,439 --> 00:01:23,519
who do decision maker tracks, so things around governance policy

25
00:01:23,560 --> 00:01:26,200
and you know how to how to manage and your

26
00:01:26,200 --> 00:01:29,480
production operations keep them secure. So it's an amazing group

27
00:01:29,519 --> 00:01:31,040
of speakers, really excited about it.

28
00:01:31,120 --> 00:01:33,560
Speaker 2: And I think I can count myself among the group

29
00:01:33,599 --> 00:01:34,599
of speakers there.

30
00:01:35,040 --> 00:01:37,120
Speaker 3: Well, yes you can. That is great.

31
00:01:37,480 --> 00:01:42,159
Speaker 2: Yeah, I'm doing a securing Blazer Server applications talk and

32
00:01:42,239 --> 00:01:45,959
also I think we're doing a Security this Week live

33
00:01:46,040 --> 00:01:48,480
show there somewhere that is correct.

34
00:01:48,719 --> 00:01:51,200
Speaker 3: Yeah, we'll be recording Security this Week Live. We're going

35
00:01:51,239 --> 00:01:54,719
to have a great panel with some folks. The interesting

36
00:01:54,719 --> 00:01:57,920
thing here is we don't really have a Microsoft and

37
00:01:58,000 --> 00:02:02,120
dot net and Azure focused toecurity conference yet, so that's

38
00:02:02,159 --> 00:02:05,159
the reason we're putting this on as well. You know

39
00:02:05,200 --> 00:02:08,280
there are other security conferences, but they have a spread

40
00:02:08,280 --> 00:02:10,439
of topics that maybe don't focus on the things you

41
00:02:10,520 --> 00:02:13,080
do day to day. And you know this overlaps with

42
00:02:13,199 --> 00:02:17,560
again our community of folks that specialize in again dot net,

43
00:02:17,680 --> 00:02:21,039
Azure and yeah, they need to keep it secure too.

44
00:02:21,280 --> 00:02:22,520
So with tons of.

45
00:02:22,520 --> 00:02:26,840
Speaker 1: Talks, cyber Intersection is part of a trio of conferences

46
00:02:26,840 --> 00:02:29,479
we're doing. They have Intersection alongside the Next Gen AI

47
00:02:29,560 --> 00:02:34,479
Conference all in Orlando the week of October fifth through tenth.

48
00:02:34,479 --> 00:02:38,039
That's workshops and the main conference. And you can get

49
00:02:38,039 --> 00:02:40,800
a special registration code if you sign up through Cybersecurity

50
00:02:40,919 --> 00:02:42,360
Intersection dot com.

51
00:02:42,680 --> 00:02:47,400
Speaker 3: Yeah, so if you sign up at Cybersecurity Intersection dot com,

52
00:02:47,599 --> 00:02:52,520
then you put in this code, so Alliance cyber three

53
00:02:52,599 --> 00:02:56,319
hundred and you'll get three hundred off the entry price.

54
00:02:56,560 --> 00:02:59,680
So that's a special code that only works at cybersecurity

55
00:03:00,599 --> 00:03:04,439
dot com. And then you have access to all the conferences.

56
00:03:04,479 --> 00:03:09,400
Speaker 2: Like Richard said, Wow, that's cool. Thanks Michelle. I'm looking

57
00:03:09,439 --> 00:03:24,120
forward to it and I'll see you there. Hey, guess what,

58
00:03:24,199 --> 00:03:26,560
it's dot net rocks all over again.

59
00:03:27,080 --> 00:03:29,199
Speaker 1: I'm Carl Franklin, an amateur Campbell.

60
00:03:29,360 --> 00:03:33,120
Speaker 2: We're at episode nineteen hundred and sixty nine.

61
00:03:33,520 --> 00:03:36,120
Speaker 1: The first time I've looked at history and thought we

62
00:03:36,199 --> 00:03:38,639
probably should do a geek out all by itself because

63
00:03:38,639 --> 00:03:40,879
this is all by itself on this year. It's the

64
00:03:41,039 --> 00:03:43,039
craziest year, just out of it.

65
00:03:43,120 --> 00:03:46,840
Speaker 2: Yeah, completely madness, cultural shift, end of the sixties. It's

66
00:03:46,960 --> 00:03:50,039
a big deal, no kidding. It was a pivotabal moment. Yeah,

67
00:03:50,120 --> 00:03:51,159
well we might as well do that.

68
00:03:51,280 --> 00:03:52,599
Speaker 1: Now you want to go right into it.

69
00:03:52,680 --> 00:03:54,680
Speaker 2: Yeah, what happened in nineteen sixty nine. I'm going to

70
00:03:54,800 --> 00:03:58,479
let you talk about the space stuff because it's pretty significant.

71
00:03:59,000 --> 00:04:00,080
Why don't you start with that?

72
00:04:00,240 --> 00:04:02,400
Speaker 1: Yeah? Sure, I mean we're talking about the Moon landing.

73
00:04:02,520 --> 00:04:06,879
So Apollo nine, which tested the lunar module in lower orbit,

74
00:04:07,080 --> 00:04:10,159
Apollo ten, which flew all the way to the Moon,

75
00:04:10,639 --> 00:04:13,560
practiced the landing, got within fifteen kilometers of the surface,

76
00:04:13,599 --> 00:04:16,759
and then aborted to test the abort systems. And then

77
00:04:17,480 --> 00:04:20,360
in July of nineteen sixty nine, the Apollo landing A

78
00:04:20,519 --> 00:04:25,720
follow eleven and Neil Armstrong and that other guy Buzz

79
00:04:26,000 --> 00:04:30,480
still good old Buzz Buzz Buzz, who was also like

80
00:04:30,519 --> 00:04:34,800
the guy He's the one who said magnificent desolation describing

81
00:04:34,839 --> 00:04:39,639
the Moon, and they pulled off this remarkable mission again

82
00:04:39,839 --> 00:04:45,000
at ridiculously high risk. Yes, that vehicle, the lunar lawn

83
00:04:45,160 --> 00:04:48,160
module was the limiting factor. It could support two people

84
00:04:48,160 --> 00:04:50,319
for three days, but it took more than three days

85
00:04:50,319 --> 00:04:54,560
to stage your rescue, So anything had failed anywhere in

86
00:04:54,560 --> 00:04:59,439
that vehicle, it was not survivable. Neil happened to bump

87
00:04:59,800 --> 00:05:03,279
the a breaker on his way out of the lunar

88
00:05:03,319 --> 00:05:05,639
module and broke that breaker. That breaker happened to be

89
00:05:05,680 --> 00:05:10,759
the power connection for the ascent engine. Ouch turned out

90
00:05:10,800 --> 00:05:13,000
that the shape of the breaker cap that would have

91
00:05:13,040 --> 00:05:15,240
pinned it back down was exactly the same shape as

92
00:05:15,319 --> 00:05:19,279
a felt pen cap, which Buzz happened to have found

93
00:05:19,519 --> 00:05:21,240
jam did in there, and that's the only reason they

94
00:05:21,240 --> 00:05:23,319
were able to get back. That's so cool. I'm sure

95
00:05:23,319 --> 00:05:25,199
they would have come up with another solution.

96
00:05:25,079 --> 00:05:27,720
Speaker 2: Well, and Apollo of their team would come later. And

97
00:05:28,399 --> 00:05:31,720
that was even so much innovation in order to get

98
00:05:31,720 --> 00:05:34,319
those guys on so much of emergency to find it,

99
00:05:34,399 --> 00:05:38,160
find a way to survive. Not to ignore the Soviets,

100
00:05:38,199 --> 00:05:40,959
but they were behind at that point. They did their

101
00:05:41,000 --> 00:05:47,360
first in orbit rendezvous that same year, but the real

102
00:05:47,360 --> 00:05:49,920
accomnchihment with the first successful landing on Venus with the

103
00:05:50,000 --> 00:05:53,279
Venera six mission that made it to the surface of Venus,

104
00:05:53,279 --> 00:05:58,639
sent back footage for about twenty minutes and then milting it.

105
00:06:00,720 --> 00:06:04,399
Other aerospace news again to ninety sixty ninety Insane seven

106
00:06:04,480 --> 00:06:08,920
forty seven's first test flight and first commercial flight in

107
00:06:09,040 --> 00:06:13,079
the same year, go Boeing also Concord's first test flight.

108
00:06:13,160 --> 00:06:16,120
Speaker 1: Nineteen sixty nine. It's crazy, but for all of us

109
00:06:16,519 --> 00:06:20,040
being computer people and what you are doing right now.

110
00:06:20,800 --> 00:06:23,519
This was the year that arpanet was turned on for

111
00:06:23,560 --> 00:06:26,439
the very first time. So this was a packet switching

112
00:06:26,480 --> 00:06:30,839
network precursor to TCPIP was all about decentralization, no central help,

113
00:06:30,959 --> 00:06:34,720
multiple routing routes. Although the first messages attempt a first

114
00:06:34,720 --> 00:06:37,319
message attempted to be transmit across the network got as

115
00:06:37,319 --> 00:06:42,279
far as L and O in login before crashing. Yeah,

116
00:06:42,639 --> 00:06:45,480
still work needed to be done. But all of that

117
00:06:46,199 --> 00:06:47,000
in nineteen.

118
00:06:46,759 --> 00:06:50,360
Speaker 2: And of course, in terms of culture, the Beatles' last

119
00:06:50,399 --> 00:06:54,399
public performance on the roof of Apple Records on January thirtieth,

120
00:06:55,240 --> 00:06:58,439
And did you watch Get Back? Yes, the remake of

121
00:06:59,439 --> 00:07:04,720
it so a great movie. Great movie and the way

122
00:07:04,759 --> 00:07:06,720
that they cleaned up the footage and everything. It was

123
00:07:06,800 --> 00:07:09,959
so much better than the quote unquote Let It Be movie,

124
00:07:09,959 --> 00:07:15,639
which was just horrible quality. Woodstock, the Woodstock Festival in

125
00:07:15,720 --> 00:07:21,879
August in New York, Upstate New York was a big deal.

126
00:07:22,079 --> 00:07:22,879
Speaker 1: Yep. Yeah.

127
00:07:23,480 --> 00:07:29,519
Speaker 2: In politics, Vietnam War escalated, significant anti war protests occurring

128
00:07:30,399 --> 00:07:37,839
across the US. The Libyan coup September one, Marmarga Daffi

129
00:07:37,959 --> 00:07:43,879
ousted King Idris the first ho Chi Minh died September second,

130
00:07:43,920 --> 00:07:48,040
at the age of seventy nine. And there was a

131
00:07:48,079 --> 00:07:51,480
few other things. But wow, what a year. Yeah, crazy years,

132
00:07:51,519 --> 00:07:54,439
extraordinary year. And we were two, so we were just

133
00:07:54,839 --> 00:07:57,199
becoming conscious of everything around us.

134
00:07:57,279 --> 00:08:00,439
Speaker 1: Not really, yeah, barely. Oh they lift off of Apollo

135
00:08:00,439 --> 00:08:03,879
eleven was July sixteenth, my birthday. Yeah. Oh so I

136
00:08:03,920 --> 00:08:05,800
turned two as the rocket was taking.

137
00:08:05,680 --> 00:08:08,920
Speaker 2: That is so cool. Random yeah, random, but very cool.

138
00:08:10,160 --> 00:08:13,759
All right, so let's do better? No framework roll the music?

139
00:08:13,800 --> 00:08:22,519
Speaker 1: Awesome? All right, man, what do you got?

140
00:08:22,800 --> 00:08:28,040
Speaker 2: I found this really cool trending repo on GitHub. Web

141
00:08:28,040 --> 00:08:32,240
gooat to web gooat a deliberately insecure web application.

142
00:08:32,480 --> 00:08:33,279
Speaker 1: Oh nice.

143
00:08:33,399 --> 00:08:38,240
Speaker 2: It's maintained by OASP designed to teach web application security lessons.

144
00:08:39,039 --> 00:08:42,159
Big disclaimers while running.

145
00:08:41,960 --> 00:08:43,200
Speaker 1: Do not deploy this.

146
00:08:43,960 --> 00:08:46,559
Speaker 2: Don't even be on the internet when you're running it.

147
00:08:46,679 --> 00:08:50,480
Oh wow, it's a demonstration of common service side application flaws.

148
00:08:50,759 --> 00:08:50,879
Speaker 1: Right.

149
00:08:51,000 --> 00:08:53,200
Speaker 2: The exercises are intended to be used by people to

150
00:08:53,320 --> 00:08:57,279
learn about application security and pen testing techniques, and so

151
00:08:57,440 --> 00:09:00,279
warning one is, while running this program, your machine will

152
00:09:00,320 --> 00:09:03,960
be extremely vulnerable to attack. You should disconnect from the

153
00:09:03,960 --> 00:09:08,120
Internet while using this program. Webgoats default configuration binds to

154
00:09:08,200 --> 00:09:12,919
local hosts to minimize exposure. And of course this program

155
00:09:13,000 --> 00:09:15,320
is for educational purposes only.

156
00:09:15,720 --> 00:09:15,879
Speaker 1: Right.

157
00:09:16,440 --> 00:09:20,200
Speaker 2: If you attempt these techniques without authorization, you are very

158
00:09:20,480 --> 00:09:23,720
likely to get caught. If you are caught engaging in

159
00:09:23,919 --> 00:09:28,159
unauthorized hacking, most companies will fire you, claiming that you

160
00:09:28,200 --> 00:09:31,039
were doing security research. Will not work? Is that as

161
00:09:31,080 --> 00:09:34,120
the first thing that all hackers claim? How about that?

162
00:09:34,320 --> 00:09:37,759
Speaker 1: Yep, don't do it. I mean, you know we're big

163
00:09:37,799 --> 00:09:40,440
on Troy hunts. You know, pen test yourself, you know,

164
00:09:40,559 --> 00:09:43,360
hack yourself, but do it with permission. Let people know

165
00:09:43,399 --> 00:09:45,919
what you're doing. You know, your intent should be good,

166
00:09:46,639 --> 00:09:47,879
be very careful.

167
00:09:47,600 --> 00:09:50,240
Speaker 2: And you know, like they say, don't be on the internet,

168
00:09:50,360 --> 00:09:53,480
don't expose yourself. Yeah, you can do all this without

169
00:09:53,879 --> 00:09:54,879
added risk.

170
00:09:54,960 --> 00:09:58,720
Speaker 1: There are tools sweeping ips all the time looking for vulnerabilities.

171
00:09:58,759 --> 00:10:01,559
You will not be it will not be long. Yeah,

172
00:10:01,600 --> 00:10:04,759
and that's it. Who's talking to us today? Richard Grabbing

173
00:10:04,799 --> 00:10:06,799
comment off a show with nineteen fifty four, the one

174
00:10:06,799 --> 00:10:09,200
we did a build with our friend w O'Brian talking

175
00:10:09,240 --> 00:10:12,919
about how AI has come to playwright with the playwright

176
00:10:13,039 --> 00:10:16,480
MCP which I think you and I both really enjoyed. Yeah,

177
00:10:16,519 --> 00:10:19,080
this is a funny comment. This is from Karthik VK

178
00:10:19,279 --> 00:10:21,639
who said in the podcast was mentioned that Microsoft should

179
00:10:21,639 --> 00:10:25,000
be leading the agent space, but I argue they already were,

180
00:10:25,320 --> 00:10:28,200
just without getting recognitional rewards. Microsoft has consistently been in

181
00:10:28,240 --> 00:10:30,440
first areas but rarely reaps the benefits. I don't know

182
00:10:30,440 --> 00:10:32,080
if I agree with you on this, Karthik, but let's

183
00:10:32,080 --> 00:10:35,480
see your argument. Take Copilot studio. It's a solid platform

184
00:10:35,519 --> 00:10:38,440
for building agents with real finesse. Semantic Kernel is another

185
00:10:38,519 --> 00:10:41,480
underrated gem, true enough, not easy to work with, but

186
00:10:41,559 --> 00:10:45,519
pretty powerful, letting developers convert existing applications into LLLM powered

187
00:10:45,559 --> 00:10:49,279
ones just by adding attributes using function calling in a

188
00:10:49,279 --> 00:10:52,440
well architected way. This is new for Microsoft. They were

189
00:10:52,480 --> 00:10:54,759
first with a touch based OS, but never got credit.

190
00:10:55,159 --> 00:10:58,000
That's definitely not true, you know, doubt They did a

191
00:10:58,039 --> 00:11:01,440
lot of experiments and built tablets early on and back

192
00:11:01,440 --> 00:11:03,799
in the XPE and so forth, but there were touch

193
00:11:03,840 --> 00:11:06,440
based interfaces. Heck, we talked about it on the show

194
00:11:06,440 --> 00:11:08,480
here going back to the sixties. So yeah, they've been

195
00:11:08,559 --> 00:11:11,440
around for a while. The much criticized Vista layout is

196
00:11:11,440 --> 00:11:13,720
now being embraced by Apple as the foundation for AR

197
00:11:13,759 --> 00:11:16,000
glasses of spatial interface. Yeah. I don't think Apple would

198
00:11:16,039 --> 00:11:20,360
coin it that way. Yeah. Wow. Yeah, the basic idea

199
00:11:20,360 --> 00:11:23,279
of bigger icons that give space for those kinds of interfaces.

200
00:11:23,320 --> 00:11:24,960
I don't know that you can copyright any of that. Yeah.

201
00:11:25,000 --> 00:11:27,879
Microsoft often builds foundational tech that shapes and ecosystem, but

202
00:11:27,960 --> 00:11:32,399
not always ways and build bring them glory. I think

203
00:11:32,720 --> 00:11:36,840
like all companies, Microsoft does allow experiments to happen and

204
00:11:36,879 --> 00:11:38,840
sometimes put them in the field, and sometimes they're too

205
00:11:38,840 --> 00:11:41,080
ahead of the market. You know, Apple may have built

206
00:11:41,120 --> 00:11:44,360
the iPhone, but they also built the Newton ten years before.

207
00:11:44,519 --> 00:11:47,919
Speaker 2: Yeah, you can't say always this and never that. I mean,

208
00:11:48,080 --> 00:11:50,919
it's just not the way it works that. Microsoft has

209
00:11:50,960 --> 00:11:53,440
made some great contributions to tech over the years, and

210
00:11:53,480 --> 00:11:56,919
also some flops. So and so is Apple. It's just

211
00:11:57,039 --> 00:11:57,360
not a.

212
00:11:57,600 --> 00:12:00,320
Speaker 1: I'm really disappointed to Courier tablet never shipped. You know,

213
00:12:00,519 --> 00:12:02,759
they they got to final prototype on that one before

214
00:12:02,759 --> 00:12:04,360
they pulled the plug on, which is two bad because

215
00:12:04,360 --> 00:12:06,399
it's it looked like an interesting machine. I not think

216
00:12:06,440 --> 00:12:08,519
that it would have succeeded, but I would have bought

217
00:12:08,519 --> 00:12:10,120
one I would have taken one up for a spin

218
00:12:10,200 --> 00:12:14,559
for sure. But yeah, Karthik, I think there's more research

219
00:12:14,639 --> 00:12:16,159
to be done if you want to see these different things.

220
00:12:16,159 --> 00:12:18,679
But I agree there's many technologies that get put out

221
00:12:18,720 --> 00:12:21,200
there but are put in the in front of people

222
00:12:21,200 --> 00:12:24,279
in a way that they necessarily embrace. If Microsoft says

223
00:12:24,360 --> 00:12:27,039
sinning for anything, it's not advocating for their own stuff

224
00:12:27,080 --> 00:12:30,480
as well as they possibly could. Often they're just building

225
00:12:30,519 --> 00:12:33,200
things and it gets out there, and whether or not

226
00:12:33,399 --> 00:12:36,159
people can see what it can do is another question entirely. Heck,

227
00:12:36,240 --> 00:12:38,080
half our shows are based on Hey, did you know

228
00:12:38,120 --> 00:12:42,440
what could do this? Yeah? Right? But that being said,

229
00:12:42,480 --> 00:12:44,200
thank you so much for your comment, and a coffee

230
00:12:44,240 --> 00:12:45,679
of music Cobuy is on its way to you. And

231
00:12:45,720 --> 00:12:47,320
if you'd like a copy of music Code, I read

232
00:12:47,320 --> 00:12:49,360
a comment on the website at dot NetRocks dot com

233
00:12:49,480 --> 00:12:51,480
or on the facebooks. We publish every show there, and

234
00:12:51,480 --> 00:12:53,120
if you comment there and we're reading the show, we'll

235
00:12:53,120 --> 00:12:54,240
send you a copy of music Code.

236
00:12:54,240 --> 00:12:56,759
Speaker 2: By Before we get started with James here, I want

237
00:12:56,799 --> 00:12:59,159
to let everybody know that Jeff Fritz and I have

238
00:12:59,240 --> 00:13:02,159
a new YouTube show that we're doing in addition to

239
00:13:02,200 --> 00:13:05,360
Blazer puzzle, and we're probably going to alternate weeks, but

240
00:13:05,440 --> 00:13:09,200
it's called code It with AI. As you probably know

241
00:13:09,399 --> 00:13:13,080
and James certainly knows because he's his boss. There's a

242
00:13:13,120 --> 00:13:18,720
big mandate to do AI content for Microsoft Evangelism, you know,

243
00:13:18,799 --> 00:13:20,639
to because there's a lot of new stuff and there's

244
00:13:20,639 --> 00:13:23,679
a lot of things to understand, and so we wanted

245
00:13:23,720 --> 00:13:25,960
to take some of the stuff that he did in

246
00:13:26,120 --> 00:13:29,960
Copilot that John, which is his website of all these

247
00:13:30,000 --> 00:13:33,519
little tips and tricks for using Copilot and other things,

248
00:13:34,039 --> 00:13:36,759
and do videos exposing some of the things that we

249
00:13:36,799 --> 00:13:39,279
can do as dot net developers, not only to help

250
00:13:39,399 --> 00:13:43,440
us write code and publish software, but to incorporate AI

251
00:13:43,600 --> 00:13:49,000
into our applications. And so we started It's interesting you started.

252
00:13:49,080 --> 00:13:52,279
You talked about Playwright. We started with the playwright MC

253
00:13:53,240 --> 00:13:57,039
was it MCP? Yeah, yeah, we started with the Playwright

254
00:13:57,200 --> 00:14:02,200
MCP to create documentation for Copilot dot com and we

255
00:14:02,240 --> 00:14:07,240
did an individual studio code with the Sonnet four to

256
00:14:07,279 --> 00:14:12,600
oh and it was amazing. It basically was a very

257
00:14:12,639 --> 00:14:16,480
small prompt and it just created a user manual for

258
00:14:16,639 --> 00:14:21,399
Copilot dot John using playwright. So that's coded with AI

259
00:14:21,519 --> 00:14:24,080
dot com. If you want to check it out and

260
00:14:24,279 --> 00:14:26,840
we will do more and let us know.

261
00:14:27,320 --> 00:14:27,720
Speaker 1: That's it.

262
00:14:27,879 --> 00:14:32,360
Speaker 2: So let's bring James on. James Montemagne is an old

263
00:14:32,360 --> 00:14:36,759
friend of ours, a developer community lead at Microsoft focused

264
00:14:36,799 --> 00:14:41,240
on building community around and helping developers learn and adopt

265
00:14:41,320 --> 00:14:46,320
the latest frameworks, languages, and agentic developer tools. Hey man,

266
00:14:46,399 --> 00:14:46,919
what's up.

267
00:14:47,120 --> 00:14:47,279
Speaker 1: Yeah?

268
00:14:47,320 --> 00:14:49,159
Speaker 4: I think that new one at the very end is

269
00:14:49,159 --> 00:14:50,600
the first time we got to add that on sence

270
00:14:50,639 --> 00:14:51,879
the last time I was on the pod.

271
00:14:52,360 --> 00:14:54,960
Speaker 1: So yeah, it's been a last time we were on.

272
00:14:55,000 --> 00:14:57,240
It was like zamorin Land, it was just a while ago,

273
00:14:57,399 --> 00:14:59,240
like entirely too long ago, to be clear.

274
00:15:00,240 --> 00:15:03,200
Speaker 4: I've missed you both, and I'm I'm very pleased and

275
00:15:03,240 --> 00:15:05,279
honored and humbled to be back on the pod. So

276
00:15:05,320 --> 00:15:06,320
it's really good to be back.

277
00:15:06,399 --> 00:15:07,200
Speaker 1: You have you on, man.

278
00:15:07,320 --> 00:15:09,480
Speaker 2: Let's give a little a little history here.

279
00:15:09,559 --> 00:15:10,120
Speaker 1: Oh my god.

280
00:15:10,480 --> 00:15:17,639
Speaker 2: Back in the days of Zamorin, we met James and Hardy, right, yeah, yeah,

281
00:15:17,759 --> 00:15:21,759
in Boston at the beginning of a dot net Rocks tour, right,

282
00:15:22,159 --> 00:15:27,039
and you guys were you know, zamorin and at that point,

283
00:15:27,240 --> 00:15:29,960
and we were talking about zamorin forms and all that stuff.

284
00:15:30,039 --> 00:15:32,399
You were talking about zamorin forms on the road trip

285
00:15:32,440 --> 00:15:35,120
with us, and we were just talking about dot net

286
00:15:35,159 --> 00:15:38,440
in general. I think, if my memory is correct, I.

287
00:15:38,360 --> 00:15:40,080
Speaker 1: Always had the sense that he was chucked under the

288
00:15:40,159 --> 00:15:41,960
bus in that sense, it's like, hey, get on this

289
00:15:42,000 --> 00:15:43,200
bus with these strange men.

290
00:15:43,919 --> 00:15:45,879
Speaker 4: It was it was my I believe it was my

291
00:15:46,000 --> 00:15:49,120
second month on the job, and I'm pretty sure that

292
00:15:49,159 --> 00:15:52,679
they said, Hey, you're going on Toro with these two dudes,

293
00:15:52,840 --> 00:15:56,919
or got this RV and just driving around the US

294
00:15:56,960 --> 00:15:58,720
and go ah.

295
00:15:58,759 --> 00:16:00,320
Speaker 2: And I don't know if you and Chris did it

296
00:16:00,759 --> 00:16:04,320
the same time, but you guys had built apps and

297
00:16:04,480 --> 00:16:07,639
got noticed by Microsoft. And I guess that's you know,

298
00:16:07,720 --> 00:16:10,440
how you worked your way into the organization, isn't it?

299
00:16:11,039 --> 00:16:11,639
The mobile app?

300
00:16:11,720 --> 00:16:11,879
Speaker 5: Yeah?

301
00:16:11,879 --> 00:16:14,200
Speaker 4: So I had been a professional mobile I worked at

302
00:16:14,240 --> 00:16:17,080
Cannon early on, writing printer software for them out of college,

303
00:16:17,519 --> 00:16:20,360
and then I went to PDC, which was right before Build.

304
00:16:20,399 --> 00:16:22,200
I was on big tent on campus and I got

305
00:16:22,200 --> 00:16:24,600
a Windows phone. I was a c sharp developer, fell

306
00:16:24,600 --> 00:16:28,519
in love with mobile development, got an Android device an iPhone,

307
00:16:28,600 --> 00:16:31,360
started building apps, and I got a job in Seattle.

308
00:16:32,399 --> 00:16:35,159
Moved my life up there as a mobile developer, found

309
00:16:35,240 --> 00:16:38,320
zamorin to do cross platform mobile devon dot net. Never

310
00:16:38,399 --> 00:16:41,440
looked back and that was it. I wrote an app

311
00:16:41,519 --> 00:16:43,600
that got featured for the company I worked at, Seaton.

312
00:16:43,639 --> 00:16:45,639
I got featured in Gadget and I was writing blog

313
00:16:45,679 --> 00:16:48,320
posts and doing kind of advocacy off to the side,

314
00:16:48,360 --> 00:16:50,600
and yeah, they randomly emailed me. I thought it was

315
00:16:50,639 --> 00:16:53,759
actually for like their MVP program. I didn't actually realize

316
00:16:53,759 --> 00:16:56,240
that it was to come in and interview like with

317
00:16:56,320 --> 00:16:59,519
Natt Friedman at the time, so I like win and

318
00:17:00,000 --> 00:17:02,159
course by then I knew I was interviewing. And yeah,

319
00:17:02,200 --> 00:17:05,119
it was three and a half years had xamred before

320
00:17:05,119 --> 00:17:08,799
the acquisition into Microsoft, which is which is awesome. So

321
00:17:09,000 --> 00:17:11,559
you know, I still building and publishing apps. I just

322
00:17:11,599 --> 00:17:14,680
published a brand new Blazer hybrid MAUI app to the

323
00:17:14,680 --> 00:17:17,680
app store last week, you know ndred percent Vibe Code,

324
00:17:17,880 --> 00:17:18,599
which was awesome.

325
00:17:18,680 --> 00:17:21,519
Speaker 2: So it was rad I remember Chris Hardy talking about

326
00:17:21,559 --> 00:17:23,720
his favorite app that he wrote was how many days

327
00:17:23,799 --> 00:17:24,799
until Christmas?

328
00:17:27,720 --> 00:17:29,079
Speaker 1: He's a big house fan.

329
00:17:29,720 --> 00:17:32,039
Speaker 4: Yeah, and I took his it was ioas only and

330
00:17:32,079 --> 00:17:34,000
I ported it to Android. Those are the very one

331
00:17:34,039 --> 00:17:35,559
of the very first things that I did, and I

332
00:17:35,640 --> 00:17:37,720
put that on the app Store with Chris, which is hilarious.

333
00:17:37,759 --> 00:17:41,119
Speaker 2: So yeah, yeah, funny, funny stuff, long long time ago.

334
00:17:41,160 --> 00:17:43,319
But we boy, that's where we met and we have

335
00:17:43,440 --> 00:17:44,359
been friends ever since.

336
00:17:44,880 --> 00:17:49,160
Speaker 4: Absolutely, yeah. And it's like it's so interesting the year stuff. Right,

337
00:17:49,240 --> 00:17:52,160
you think you were, you know, around in nineteen sixty nine,

338
00:17:52,200 --> 00:17:54,559
Like what a time to be alive. And that's kind

339
00:17:54,599 --> 00:17:57,079
of like now. I kind of like think that now,

340
00:17:58,000 --> 00:18:00,880
and just like how everything is. Everything's always be moving fast, right,

341
00:18:00,920 --> 00:18:04,240
but it feels like things are super accelerated. But if

342
00:18:04,240 --> 00:18:06,759
you've latched on, I feel like it's in a really

343
00:18:07,480 --> 00:18:10,000
really fun and interesting place. I'm excited to dive in

344
00:18:10,079 --> 00:18:12,079
with kind of some new topics with y'all.

345
00:18:12,319 --> 00:18:14,599
Speaker 2: Well yeah, I mean we don't. We haven't really we

346
00:18:14,640 --> 00:18:17,000
don't really talk about visual studio code that much because

347
00:18:17,000 --> 00:18:19,599
I don't think Richard and I use it all that much.

348
00:18:19,599 --> 00:18:20,880
I don't know about you, Richard, but.

349
00:18:21,160 --> 00:18:22,680
Speaker 1: I'm in and out of it all the time. But

350
00:18:22,880 --> 00:18:26,119
you know we're we're studio people or ide people, right,

351
00:18:26,440 --> 00:18:28,599
that's where we came from. I don't know that when

352
00:18:28,599 --> 00:18:30,400
I want to develop, that's where I go. When I

353
00:18:30,440 --> 00:18:34,559
want to edit a zambal doc. I mean I mean

354
00:18:34,640 --> 00:18:35,599
studio code, but.

355
00:18:36,240 --> 00:18:38,680
Speaker 2: I've found that the agents work better in studio code

356
00:18:38,680 --> 00:18:41,440
than they do in studio probably because of the threading

357
00:18:41,480 --> 00:18:43,599
model or something like that. I don't know what's going on,

358
00:18:43,680 --> 00:18:47,799
but I really really enjoy it. You know, every time

359
00:18:47,839 --> 00:18:50,759
that Fritz and I do something in studio code, I'm like, hmm,

360
00:18:51,119 --> 00:18:55,160
maybe maybe I'll you know, although I have customers that

361
00:18:55,240 --> 00:18:57,240
are in studio and I have to use that, so.

362
00:18:57,440 --> 00:18:59,759
Speaker 4: Yeah, I think I think, you know, for me, it

363
00:18:59,839 --> 00:19:03,920
was kind of in maybe January or February this year

364
00:19:04,039 --> 00:19:06,680
or I kind of made this leap and jump, and

365
00:19:06,720 --> 00:19:08,599
I think a lot of developers are early on, like

366
00:19:08,680 --> 00:19:12,039
in their sort of like how much AI coding stuff

367
00:19:12,079 --> 00:19:14,480
do I adopt every single day and their journey, Like

368
00:19:14,480 --> 00:19:16,799
we kind of think that everybody's using it, but that's

369
00:19:16,839 --> 00:19:19,920
not the case. However, many people are and adopting it

370
00:19:20,000 --> 00:19:22,519
kind of slowly. But we have always been if you

371
00:19:22,519 --> 00:19:25,200
think about just intelli Sense and intelecode, and then we

372
00:19:25,240 --> 00:19:27,359
have the extensions that have been giving us and helping

373
00:19:27,400 --> 00:19:28,440
us write code faster.

374
00:19:29,079 --> 00:19:30,079
Speaker 1: So really, in.

375
00:19:30,079 --> 00:19:33,400
Speaker 4: January February, I kind of did dive all in when

376
00:19:33,440 --> 00:19:37,519
agent mode dropped inside of VS code and I really

377
00:19:37,960 --> 00:19:40,359
dove in deeper because the c sharp Defocate was getting better,

378
00:19:40,640 --> 00:19:43,480
the Maui extensions were getting better, the Plazer integrations were

379
00:19:43,480 --> 00:19:46,359
getting better, and it's really just dove all in. And

380
00:19:46,400 --> 00:19:48,400
I like to say I like gave in. I gave

381
00:19:48,400 --> 00:19:52,759
myself to agent mode, and you know, I've gone back

382
00:19:52,799 --> 00:19:54,759
and forth kind of talking about the IDEs, like the

383
00:19:54,880 --> 00:19:58,519
vs team has been doing great, like you know, adding

384
00:19:58,559 --> 00:20:00,759
more and more features to the Relief twenty twenty six

385
00:20:00,880 --> 00:20:04,200
is coming out soon, and like there's more and more integration.

386
00:20:04,319 --> 00:20:06,720
So it feels like they have some unique features that

387
00:20:06,799 --> 00:20:08,799
are like the profile aer and some of the debugging

388
00:20:08,880 --> 00:20:12,240
stuff for ID specific things. But yeah, the agent mode

389
00:20:12,279 --> 00:20:15,960
and the rapid pace. Like I'm on Insider for VS

390
00:20:16,000 --> 00:20:19,400
code and I'm just getting updates every single night, just

391
00:20:19,400 --> 00:20:21,759
like rapid, right, and that's how I live. I don't stable,

392
00:20:21,799 --> 00:20:23,319
I on the insider and just go.

393
00:20:24,000 --> 00:20:26,680
Speaker 1: I feel like VS code can move faster into this

394
00:20:26,680 --> 00:20:31,559
new paradigm than the studio can, Like studio customers tend

395
00:20:31,640 --> 00:20:34,599
to be supporting large projects, like we tend to not

396
00:20:34,680 --> 00:20:39,680
emphasize the studio responsibilities to project management as much as

397
00:20:39,720 --> 00:20:42,599
it is to coding space as well, Like the show

398
00:20:42,599 --> 00:20:44,119
we did with Mads a few weeks ago talking about

399
00:20:44,119 --> 00:20:46,160
Studio twenty twenty six. He's very clear this is the

400
00:20:46,200 --> 00:20:50,359
AI version of Studio and it's still coming where you know,

401
00:20:50,519 --> 00:20:53,559
Studio Code had this in the spring to some of

402
00:20:53,640 --> 00:20:56,200
the rate. Obviously there's more to be done, but it's like, hey,

403
00:20:56,440 --> 00:20:59,480
would you think's going to happen with this behemoth that

404
00:20:59,599 --> 00:21:02,079
is an eye like it only goes so quickly. They've

405
00:21:02,279 --> 00:21:04,200
I've really done the plugins and so forth, but the

406
00:21:04,200 --> 00:21:06,640
integration is not the same. It'll be interesting to see

407
00:21:06,640 --> 00:21:08,880
where they get to. But this is far from a

408
00:21:09,240 --> 00:21:10,319
played out store.

409
00:21:10,079 --> 00:21:14,119
Speaker 2: We were talking to Dustin Campbell and I was sort

410
00:21:14,160 --> 00:21:19,559
of complaining a little bit about the Razor Editor. He says, oh, yeah,

411
00:21:19,640 --> 00:21:22,960
I got that, man, I'm working on that, and apparently

412
00:21:23,000 --> 00:21:25,880
he has. I haven't seen twenty twenty six yet, but

413
00:21:26,119 --> 00:21:29,680
Fritz says he has, and the Razor Editor is like

414
00:21:29,960 --> 00:21:32,759
night and day of what it is and you know,

415
00:21:32,799 --> 00:21:34,039
currently in twenty twenty two.

416
00:21:34,359 --> 00:21:35,759
Speaker 1: Yeah, I think so, I can't wait.

417
00:21:35,880 --> 00:21:38,000
Speaker 4: The teams are really pushing super hard, and I think

418
00:21:38,160 --> 00:21:40,400
it's a good point. Like I also think that it's

419
00:21:40,440 --> 00:21:43,880
great that these you know, two paradigms exist. A lightweight

420
00:21:43,920 --> 00:21:47,119
code editor that's an AI first open source everything editor,

421
00:21:47,160 --> 00:21:50,279
and then visual Studio, which is this ide with all

422
00:21:50,279 --> 00:21:53,799
these big workloads sets up everything for you, right, And

423
00:21:54,319 --> 00:21:56,960
I think with the Visual Studio, right, it's also not

424
00:21:57,400 --> 00:21:59,720
just I'm going to open a project, it's that they're

425
00:21:59,759 --> 00:22:04,359
come and is that And individuals that have huge, crazy projects, right,

426
00:22:04,440 --> 00:22:06,960
huge c plus plus games like game studios are using

427
00:22:07,039 --> 00:22:09,000
these things with you know, hundreds of millions of lines

428
00:22:09,039 --> 00:22:12,000
of code, right, and they're like legacy projects too, and

429
00:22:12,079 --> 00:22:13,440
you have to think about how do you support the

430
00:22:13,599 --> 00:22:15,880
really really old stuff and the new stuff and then

431
00:22:15,960 --> 00:22:19,279
make all of that AI stuff work seamlessly across all

432
00:22:19,359 --> 00:22:23,279
of that. That's a big chat challenge to represent compared to, Hey,

433
00:22:23,279 --> 00:22:26,000
I'm on modern stack, right, Like I built this at

434
00:22:26,039 --> 00:22:29,079
feedback Flow, which you know, I've one hundred percent vibe

435
00:22:29,079 --> 00:22:32,920
coded AI and I went back and forth between VS

436
00:22:32,960 --> 00:22:35,880
and VS code. But that's all modern, right. It's it's

437
00:22:35,880 --> 00:22:38,799
done at nine, it's Blazer, it's Azure Functions, it's done

438
00:22:38,839 --> 00:22:41,839
at APIs, it's modern MCP stack. So I'm in the

439
00:22:41,880 --> 00:22:46,079
modern world of doing stuff inside of there. And that

440
00:22:46,200 --> 00:22:47,920
was great, you know, that I could go and I

441
00:22:47,920 --> 00:22:49,519
could also open that in VS if I need to

442
00:22:49,519 --> 00:22:52,599
do deep debugging or do like some advanced profiling or

443
00:22:52,599 --> 00:22:55,599
things like that. But I can also code anywhere. Right

444
00:22:55,640 --> 00:22:57,359
right now, I'm on my my Mac Mini, I got

445
00:22:57,400 --> 00:22:59,279
my surface loptop, I got all my devices. So that's

446
00:22:59,279 --> 00:23:01,799
where that sort of experience goes. I think it's kind

447
00:23:01,799 --> 00:23:03,559
of a It's always a great time to be alive

448
00:23:03,599 --> 00:23:06,200
as a developer because things are evolving. But just that

449
00:23:06,319 --> 00:23:08,960
choice and flexibility I think is important. And we say

450
00:23:08,960 --> 00:23:10,960
that with AI as well. There's lots of choices out there.

451
00:23:11,240 --> 00:23:14,920
Speaker 1: I'm seeing that lots of teams, especially those they also

452
00:23:14,960 --> 00:23:19,079
have younger generation developers. They are mixing the two like

453
00:23:19,279 --> 00:23:21,359
studio and studio code, you know, especially once you get

454
00:23:21,359 --> 00:23:23,559
dev kid in the equation, they work and play well together.

455
00:23:23,720 --> 00:23:26,440
And a lot of web devs and again I'm going

456
00:23:26,480 --> 00:23:29,880
to say skew younger, they're not interested in the ide

457
00:23:29,920 --> 00:23:32,839
they learned on studio code. That's how they want to develop.

458
00:23:32,880 --> 00:23:34,279
They have a plan on how they want to do that,

459
00:23:34,319 --> 00:23:37,480
but they need to work within those larger projects that

460
00:23:37,720 --> 00:23:39,960
let's face it, more senior folks are living, you know,

461
00:23:39,960 --> 00:23:42,359
we're originally built in the IDE and a lot of

462
00:23:42,400 --> 00:23:44,559
the dependency on that, so it's not like these two

463
00:23:44,599 --> 00:23:46,480
are mutually exclusive to each other now.

464
00:23:46,599 --> 00:23:48,920
Speaker 4: And I think the team's done a pretty good job,

465
00:23:49,400 --> 00:23:53,519
especially in this last year, especially as Visual Studio has

466
00:23:53,720 --> 00:23:56,720
actually adopted a faster iteration cycle instead of every quarter

467
00:23:56,799 --> 00:23:59,720
every month, and you're actually seeing a lot more parody

468
00:24:00,200 --> 00:24:03,640
jumping between the two, right actually, as far as model selection,

469
00:24:04,079 --> 00:24:07,400
how the actual like agent mode and chat modes work,

470
00:24:07,480 --> 00:24:10,559
and how different integrations like now with coding agents are

471
00:24:10,559 --> 00:24:13,240
being integrated between the two. So VS Code because it

472
00:24:13,519 --> 00:24:16,519
ships crazy fast, you know, is going to have things

473
00:24:16,680 --> 00:24:20,920
super fast first, but also VS will soon follow up. Right,

474
00:24:21,079 --> 00:24:24,279
it's going through different sort of you know, rolling out

475
00:24:24,319 --> 00:24:27,359
as just sort of people adopt things at a different pace,

476
00:24:27,640 --> 00:24:30,839
but also adding unique features, like I said, specific for

477
00:24:31,039 --> 00:24:32,440
that type of development.

478
00:24:32,000 --> 00:24:34,559
Speaker 1: Being done, and you would hope one informs the other two,

479
00:24:34,680 --> 00:24:37,279
like what they learn from those modules running in studio

480
00:24:37,319 --> 00:24:40,640
code then is reflected in studio. You know, can can

481
00:24:41,039 --> 00:24:43,359
build a better version or more it may make more

482
00:24:43,400 --> 00:24:44,680
sense for that customer.

483
00:24:44,279 --> 00:24:46,920
Speaker 4: Base, absolutely, you know, And that's why I like the naming.

484
00:24:47,039 --> 00:24:49,400
The names are the same, right, ask an agent they're

485
00:24:49,440 --> 00:24:51,680
the same. Right, It'd be really weird if you open

486
00:24:51,720 --> 00:24:54,480
the same project in both VS Code and Visual Studio

487
00:24:54,759 --> 00:24:57,200
and then like everything is one hundred percent different, right,

488
00:24:57,240 --> 00:25:00,680
So even icons and placement and things thought about. I

489
00:25:00,720 --> 00:25:05,039
think at that factor, however, like inherently they're different. They're

490
00:25:05,039 --> 00:25:09,039
different editors, they're different spaces inside of there. And for

491
00:25:09,119 --> 00:25:12,039
me at least, I've really enjoyed kind of being on

492
00:25:12,039 --> 00:25:15,960
this like super breakneck fast, you know, on the things

493
00:25:15,960 --> 00:25:17,880
that I'm building. I like to say the year of

494
00:25:17,920 --> 00:25:20,799
twenty twenty five was the year that I shipped and

495
00:25:20,839 --> 00:25:25,240
wrote more code in my entire life, at multiple levels

496
00:25:25,240 --> 00:25:29,359
small like little prototype levels, to large production applications that

497
00:25:29,440 --> 00:25:32,799
are infusing you know, AI elements of foundry into them,

498
00:25:32,960 --> 00:25:37,119
to functions to databases, and nearly none of the code

499
00:25:37,160 --> 00:25:39,720
I wrote by hand at all. Right, I'm actually like, really,

500
00:25:40,000 --> 00:25:42,240
like I said, dove into this prompt first type of

501
00:25:42,279 --> 00:25:47,160
development between both VS Code and VS and it's really

502
00:25:47,200 --> 00:25:51,160
fascinating to watch the editors evolve and also the deep

503
00:25:51,240 --> 00:25:54,599
understanding that like VS and VS Code and Microsoft itself

504
00:25:54,599 --> 00:25:57,640
in the developer space, you know, developers, right, So we're

505
00:25:57,680 --> 00:25:59,960
building tools for developers and we're dog footing, right.

506
00:26:00,039 --> 00:26:00,880
Speaker 1: The vs Code.

507
00:26:00,640 --> 00:26:03,839
Speaker 4: Team builds vs code with vs code and agent mode

508
00:26:03,839 --> 00:26:05,720
and these things, and same with the Visual Studio team.

509
00:26:05,799 --> 00:26:08,119
So it's like deep dog fooding and understand that we've

510
00:26:08,119 --> 00:26:11,680
been doing this for twenty years, thirty years, whatever it

511
00:26:11,720 --> 00:26:14,960
is now that deep understanding of how software is built.

512
00:26:14,960 --> 00:26:17,480
Speaker 2: Can we talk about the models a little bit. It's

513
00:26:17,640 --> 00:26:20,240
my understanding that Claude's on it for is like the

514
00:26:20,279 --> 00:26:28,519
best right now for coding C sharp, Blazer, CSS, JavaScript.

515
00:26:28,920 --> 00:26:30,200
What's your take on that?

516
00:26:30,319 --> 00:26:38,680
Speaker 4: Well, Carl, back in my day in February weeks ago,

517
00:26:39,319 --> 00:26:41,119
you know it's now fun because you know all the

518
00:26:41,119 --> 00:26:43,759
web devs like with JavaScript lies. Oh it's a new day,

519
00:26:43,759 --> 00:26:45,200
it's a new like. Now it's a new model.

520
00:26:45,240 --> 00:26:45,440
Speaker 1: Right.

521
00:26:45,640 --> 00:26:47,759
Speaker 4: So when I always when I super dove in like

522
00:26:47,839 --> 00:26:51,640
Clauds on it, three five Droma and four is using there,

523
00:26:52,119 --> 00:26:54,519
I think there's a few things models or models will

524
00:26:54,519 --> 00:26:57,000
be new models they have all the time. I think

525
00:26:57,039 --> 00:27:00,079
for me, it's it's when I think of this and

526
00:27:00,079 --> 00:27:02,160
if I was to encouraged developers listening, it's a new

527
00:27:02,240 --> 00:27:04,480
tool in your toolbox. Every model is a new tool.

528
00:27:04,559 --> 00:27:08,160
Inside of that toolbox is ask, which is agent mode,

529
00:27:08,160 --> 00:27:11,200
which is coding agents that are working autonomous in the background.

530
00:27:11,680 --> 00:27:14,599
For me, it's a great question. It comes down to

531
00:27:15,680 --> 00:27:17,440
how do you want to work and how do you

532
00:27:17,440 --> 00:27:18,880
want to work with your model? So let me break

533
00:27:18,880 --> 00:27:20,480
it down into two categories.

534
00:27:20,720 --> 00:27:21,440
Speaker 1: You kind of have.

535
00:27:22,000 --> 00:27:25,079
Speaker 4: You have the GPT models and I usually am in

536
00:27:25,200 --> 00:27:29,680
GPT five Mini or GBT five and a lot of Sonnet.

537
00:27:29,880 --> 00:27:31,400
I go back and forth, and I'll tell you why

538
00:27:31,440 --> 00:27:35,000
I go back and forth. Okay, that these models inherently

539
00:27:35,599 --> 00:27:38,759
work and think different and they're different people. It is

540
00:27:38,839 --> 00:27:41,880
if I have two different co workers sitting side by

541
00:27:41,920 --> 00:27:42,880
side of me, working.

542
00:27:42,680 --> 00:27:45,920
Speaker 1: With different kinds of brain damage, basically.

543
00:27:45,599 --> 00:27:49,559
Speaker 4: All sorts of different thinking and logic and type of

544
00:27:49,559 --> 00:27:51,799
code that they write. So I think with the GPT

545
00:27:51,960 --> 00:27:55,799
models they like to kind of be told what to do,

546
00:27:56,240 --> 00:27:58,720
like what files, what do you want to work on?

547
00:27:59,240 --> 00:28:00,960
You know, how do you want to work on it?

548
00:28:01,039 --> 00:28:03,839
And go off and do it. They are very much

549
00:28:04,400 --> 00:28:08,519
give me a ticket, describe the specs. I gotcha, right.

550
00:28:08,559 --> 00:28:10,480
They're very very good at that, and they're very very

551
00:28:10,519 --> 00:28:12,960
fast at it, right to be more pointed at it,

552
00:28:13,559 --> 00:28:15,920
and that's good in a lot of scenarios like bug fixing,

553
00:28:16,000 --> 00:28:20,039
like examining, just looking and doing ask and like kind

554
00:28:20,039 --> 00:28:22,640
of getting detailed information quickly, because that's.

555
00:28:22,720 --> 00:28:26,720
Speaker 2: That's what the GitHub agent uses, right, getub agent, the

556
00:28:26,880 --> 00:28:29,720
on GitHub, the coding agent, the coding Yeah, the coding agent.

557
00:28:29,839 --> 00:28:32,359
Speaker 1: I think it bops between a few models, does it? Yeah?

558
00:28:32,440 --> 00:28:33,759
I think it does fast.

559
00:28:33,799 --> 00:28:36,599
Speaker 2: I checked it GPT only, but the son it was coming,

560
00:28:36,640 --> 00:28:37,759
I guess yeah.

561
00:28:37,759 --> 00:28:39,559
Speaker 4: For a while it was just son it for I

562
00:28:39,599 --> 00:28:41,640
think so. Really, yeah, I think so.

563
00:28:42,839 --> 00:28:43,400
Speaker 1: But son it.

564
00:28:43,440 --> 00:28:50,759
Speaker 4: Models are fascinating. They are super curious, and they are ambitious,

565
00:28:51,400 --> 00:28:56,119
and they take time to understand a lot of the context,

566
00:28:56,359 --> 00:29:00,680
explore the code base. And when you tell it to

567
00:29:00,759 --> 00:29:02,599
do something and you ask it and you kind of

568
00:29:02,640 --> 00:29:05,440
give you know, smaller, medium, or even large prompts or

569
00:29:05,440 --> 00:29:08,880
assign an issue, they like to get in there. They

570
00:29:08,920 --> 00:29:12,200
like to explore, right, They like to just just figure

571
00:29:12,240 --> 00:29:14,759
out all the little nooks and crannies and what qutblem

572
00:29:15,240 --> 00:29:17,640
break it down. And what claud will do though, is

573
00:29:17,799 --> 00:29:20,240
it will do things you don't necessarily maybe even want

574
00:29:20,279 --> 00:29:22,039
it to do. But then you're like, maybe I did

575
00:29:22,039 --> 00:29:22,720
want it to do that.

576
00:29:22,799 --> 00:29:23,240
Speaker 1: I don't know.

577
00:29:23,480 --> 00:29:25,440
Speaker 4: It'll start updating things. It's like, oh, I have to

578
00:29:25,440 --> 00:29:27,039
dis method. Oh, I should update the docs or I

579
00:29:27,079 --> 00:29:30,680
should do this right, And as it sort of context grows,

580
00:29:31,240 --> 00:29:33,839
it will start to like really explore the COVID, which

581
00:29:33,880 --> 00:29:37,880
is good and bad because it's good and that it

582
00:29:37,920 --> 00:29:41,079
may you know, get things that you missed, but also

583
00:29:41,119 --> 00:29:43,119
at the same time, it takes longer. Right, you just

584
00:29:43,160 --> 00:29:44,759
could be letting it churn and kind of letting it

585
00:29:44,799 --> 00:29:48,759
do stuff which background coding agents, like background tasks, background agents,

586
00:29:48,799 --> 00:29:52,759
things like that that run autonomously out there. That's great

587
00:29:52,799 --> 00:29:54,279
because they can take a lot of time. They can

588
00:29:54,319 --> 00:29:56,480
be very verbose to get run tests. But I've seen

589
00:29:56,480 --> 00:29:58,279
with Claude, for example, it's like, hey, let me write

590
00:29:58,279 --> 00:29:59,839
this tone Okay, let me run this test. I'm gonna

591
00:29:59,839 --> 00:30:01,759
write test, I'm gonna write the docs. I'm gonna run this,

592
00:30:01,759 --> 00:30:03,160
and I'm gonna run this, and you're gonna run this.

593
00:30:03,200 --> 00:30:05,519
You're like, Wow, you just did an entire test suite

594
00:30:05,559 --> 00:30:05,839
and all.

595
00:30:05,680 --> 00:30:09,960
Speaker 1: I can do is an overachiever. Uh employee.

596
00:30:10,119 --> 00:30:12,759
Speaker 4: It's ambitious. Yeah, they're really ambitious, and you want that

597
00:30:12,799 --> 00:30:14,279
sometimes and sometimes you don't.

598
00:30:14,359 --> 00:30:16,799
Speaker 1: I've always got the sense that claud it's like they're

599
00:30:16,839 --> 00:30:20,400
pre the prompt you write to chet GPT is the

600
00:30:20,440 --> 00:30:22,759
prompt that arrives at chet GPG. When I write a

601
00:30:22,759 --> 00:30:25,000
prompt to Claude, it's like somebody added a bunch of

602
00:30:25,000 --> 00:30:27,240
that stuff, a bunch of stuff to the prompt to

603
00:30:27,400 --> 00:30:28,000
do more.

604
00:30:28,759 --> 00:30:30,599
Speaker 4: So there's a few things, you know, I think of

605
00:30:30,640 --> 00:30:32,759
best practices here of how do we get these models

606
00:30:32,759 --> 00:30:35,480
to generate code as if we were writing it. One

607
00:30:35,599 --> 00:30:37,680
is like the team, the VS and VS code team.

608
00:30:37,720 --> 00:30:39,759
Like when you send a prompt, there is a system

609
00:30:39,799 --> 00:30:42,640
prompt that also gets sent, right because there's tools, there's mcps, results,

610
00:30:42,640 --> 00:30:45,559
other stuff. Each model has its own prompt, right because

611
00:30:45,559 --> 00:30:48,599
each model is different. So the team is working directly

612
00:30:48,880 --> 00:30:52,079
with Anthropic and open AI and these other model vendors

613
00:30:52,079 --> 00:30:54,000
to make sure their models work great based on how

614
00:30:54,079 --> 00:30:56,319
they built a model. But then there's stuff that you

615
00:30:56,359 --> 00:30:59,119
can do right. So for example, agents dot MD and

616
00:30:59,200 --> 00:31:02,799
copilot and diductions, which are instruction files that get sent

617
00:31:02,920 --> 00:31:06,319
with every single request that you put. So think of

618
00:31:06,359 --> 00:31:08,519
it as your team's best practices. How do you want

619
00:31:08,519 --> 00:31:11,200
your code generated? Do you want your CSS and eraser

620
00:31:11,240 --> 00:31:13,799
dot CSS or do you want to interact dot CSS.

621
00:31:13,960 --> 00:31:15,880
Do you want things to be light theme and dark theme?

622
00:31:15,960 --> 00:31:21,519
Do you want specific you know, M underscore, underscore, S underscore, CamelCase,

623
00:31:21,599 --> 00:31:24,480
Pascal case, how do you code? The model can infer,

624
00:31:25,200 --> 00:31:28,759
but the model also wants to please, and it wants

625
00:31:28,759 --> 00:31:31,759
to please quickly. Right, So like if you think of GPT,

626
00:31:31,920 --> 00:31:36,039
especially for one or five five mini that aren't necessarily

627
00:31:36,240 --> 00:31:39,200
deep thinking models, they want to respond to you as

628
00:31:39,240 --> 00:31:40,480
fast as humanly possible.

629
00:31:40,559 --> 00:31:43,319
Speaker 2: Right, So, if you're right, they want to be Promorphization

630
00:31:43,519 --> 00:31:46,000
is killing Richard, I can just well, no, don't.

631
00:31:45,880 --> 00:31:49,440
Speaker 4: Worry, it's coming. You're absolutely correct. You're absolutely correct. I

632
00:31:49,440 --> 00:31:53,680
see the mistake now, you're right, and they want to

633
00:31:53,680 --> 00:31:56,839
make you they want they want to make you happy inherently, right.

634
00:31:57,000 --> 00:32:00,519
So that's why they have this like verbiage that is like,

635
00:32:00,559 --> 00:32:02,880
oh no, you're absolutely right, you're totally you're totally good. Yep,

636
00:32:03,200 --> 00:32:06,720
I see the problem up good? Yeah, Oh I fixed it.

637
00:32:06,799 --> 00:32:07,039
Speaker 1: Did you?

638
00:32:07,480 --> 00:32:12,440
Speaker 5: I don't know what I're looking for is obsequious, but

639
00:32:12,680 --> 00:32:14,799
you know when you think about it, you know, if

640
00:32:14,799 --> 00:32:19,200
you had another engineer sitting side by side, do you

641
00:32:19,200 --> 00:32:20,640
you know you'd be looking at the code.

642
00:32:20,799 --> 00:32:22,759
Speaker 4: Oh yeah, I do see the problem there, it is, right,

643
00:32:22,880 --> 00:32:23,400
let's fix it.

644
00:32:23,799 --> 00:32:26,279
Speaker 1: When you actually pair a program. You are pretty kind

645
00:32:26,279 --> 00:32:29,480
to each other because we've all sat in the seat, right, Yeah,

646
00:32:29,519 --> 00:32:30,799
we've all sat beside the seat.

647
00:32:31,039 --> 00:32:33,400
Speaker 4: If you look at it that way, I think that's

648
00:32:33,400 --> 00:32:35,839
the way to achieve it. And then also not giving

649
00:32:35,920 --> 00:32:38,319
up on it. When I built feedback Flow, like I said,

650
00:32:38,319 --> 00:32:40,400
it's it's hundreds of thousands of lines of code and

651
00:32:41,000 --> 00:32:46,119
nice architecture and fully open source. And I went into

652
00:32:46,240 --> 00:32:49,519
it saying I don't want to write any code. I

653
00:32:49,559 --> 00:32:54,440
want to really dive deep into understanding how every model works,

654
00:32:54,480 --> 00:32:57,079
how the agent works, how I can customize my instructions,

655
00:32:57,119 --> 00:32:58,799
how I can get this working. And I'm at the

656
00:32:58,839 --> 00:33:01,160
point now I don't even run the app on on

657
00:33:01,200 --> 00:33:03,519
my local machine. I just push it to a branch,

658
00:33:03,640 --> 00:33:06,359
do a PR, it goes into staging, have the I

659
00:33:06,359 --> 00:33:08,279
don't even run it. I don't even need to because

660
00:33:08,359 --> 00:33:11,759
it's gotten to the point that I've massaged the infrastructure

661
00:33:11,799 --> 00:33:15,559
around it so much that the thing is building it,

662
00:33:15,559 --> 00:33:17,799
it's running the test, that's doing all the stuff before

663
00:33:17,839 --> 00:33:20,000
I push the code, sure that it's either going to

664
00:33:20,000 --> 00:33:22,200
look or not right, and just wasting time running it

665
00:33:22,200 --> 00:33:24,279
and testing it's not going to pass the tests if

666
00:33:24,279 --> 00:33:24,480
it is.

667
00:33:24,799 --> 00:33:27,799
Speaker 2: When you say massaging the infrastructure, do you mean like

668
00:33:27,920 --> 00:33:30,559
setting up a context so it kind of knows your

669
00:33:30,559 --> 00:33:33,000
style and it and you said it infers it, but

670
00:33:33,079 --> 00:33:34,079
does it remember it?

671
00:33:34,759 --> 00:33:35,759
Speaker 1: Like is there?

672
00:33:35,799 --> 00:33:38,519
Speaker 2: Do the agents have enough context to learn what I

673
00:33:38,680 --> 00:33:40,079
like and keep doing it that way?

674
00:33:40,279 --> 00:33:43,079
Speaker 4: Right now you're in the mode of telling it kind

675
00:33:43,119 --> 00:33:45,119
of how you wanted to do. So that agent's dot

676
00:33:45,240 --> 00:33:48,599
mdfile or the copilot instructions filed. They're the most important

677
00:33:48,640 --> 00:33:52,680
files in any project. System prompts, yeah, they're well, they're

678
00:33:53,480 --> 00:33:57,480
they are not necessarily system prompts. Think of them as

679
00:33:58,200 --> 00:34:03,480
a set of you know, guidance that you send with

680
00:34:03,519 --> 00:34:06,519
every prompt. So for example, it'll tell it like what

681
00:34:06,559 --> 00:34:09,880
the projects are, what frameworks they are, how you like

682
00:34:09,920 --> 00:34:12,119
your CSS, how you like your c sharp, how you

683
00:34:12,159 --> 00:34:16,280
like these things, how you like different things constructed in

684
00:34:16,320 --> 00:34:19,320
your application. Maybe for example, like you prefer using XI

685
00:34:19,440 --> 00:34:21,559
nate over MS tests, and the tests are run here

686
00:34:21,559 --> 00:34:23,599
and this is how you run them, or using aspires,

687
00:34:23,639 --> 00:34:25,360
or here's how you run a spire, Here's how you

688
00:34:25,639 --> 00:34:28,559
want your CSS versus not. And with every request that

689
00:34:28,599 --> 00:34:31,599
gets sent off, so it gets attached to the system

690
00:34:31,639 --> 00:34:33,840
prompt So give it the guidance of the context. Now

691
00:34:34,280 --> 00:34:37,519
that being said, there's not like there's memories today. I

692
00:34:37,559 --> 00:34:41,000
mean you can inherently create memories. So I often have

693
00:34:41,079 --> 00:34:44,880
a docs folder or an ideas folder or you know,

694
00:34:45,039 --> 00:34:47,599
kind of something in my repo, kind of like spectruve

695
00:34:47,599 --> 00:34:49,800
in development, like here's my specification, so it could go

696
00:34:49,840 --> 00:34:52,599
look at how I want things created. But when you're

697
00:34:52,639 --> 00:34:56,719
in the agent chat, there that entire, entire context is

698
00:34:56,760 --> 00:34:58,719
being sent back and forth. Right, So the memories, if

699
00:34:58,719 --> 00:35:01,639
you're working on a big feature, you inherently are like, okay,

700
00:35:01,639 --> 00:35:03,280
I want to start new chats and I'm working on

701
00:35:03,320 --> 00:35:06,199
something different. It's actually better to keep that thing around

702
00:35:06,440 --> 00:35:09,960
until you've implemented or fixed that bug and then change context.

703
00:35:09,960 --> 00:35:12,400
It's almost like opening and closing a ticket because that

704
00:35:12,480 --> 00:35:16,840
context is there now. Ideally over time and I think

705
00:35:16,840 --> 00:35:19,880
we'll get there and probably not the far future. Is

706
00:35:19,880 --> 00:35:22,760
that these bits of memory, right Like Carl, for example,

707
00:35:22,800 --> 00:35:25,039
you're like no, no, no, I really want my CSS

708
00:35:25,079 --> 00:35:27,480
this way, blah blah blah. It should remember that, and

709
00:35:27,519 --> 00:35:29,760
it should remember that just like you would write that

710
00:35:29,840 --> 00:35:32,559
in a documentation. So right now today, what I do

711
00:35:32,679 --> 00:35:35,079
is whenever I see something wrong that it generated and

712
00:35:35,079 --> 00:35:35,639
I asked.

713
00:35:35,400 --> 00:35:35,960
Speaker 1: It to fix it.

714
00:35:36,000 --> 00:35:38,159
Speaker 4: I say, oh, and can you go write a note

715
00:35:38,159 --> 00:35:41,079
in the copilot instructions so you don't get this mistake again.

716
00:35:41,199 --> 00:35:44,880
So I'm telling it to go inherently keeping it memory

717
00:35:45,039 --> 00:35:45,480
out there.

718
00:35:45,679 --> 00:35:50,480
Speaker 2: I've found with the GitHub code agent coding agent that

719
00:35:50,639 --> 00:35:54,679
even though I did that, it's still insisted every time

720
00:35:54,719 --> 00:35:58,199
I asked it to do some laser coding to take

721
00:35:58,239 --> 00:36:02,679
my existing dot net nine application and downgrade it to

722
00:36:02,719 --> 00:36:05,840
dot net eight for set, no matter how many times

723
00:36:06,000 --> 00:36:09,280
I say, don't do this, keep storing it.

724
00:36:09,800 --> 00:36:12,519
Speaker 4: Okay, okay, okay, So here's what's happening with coding agents.

725
00:36:12,960 --> 00:36:13,880
Speaker 1: Bluish pluck here.

726
00:36:15,440 --> 00:36:17,840
Speaker 4: Okay, so I just I just watched this as black

727
00:36:17,920 --> 00:36:21,280
like whole labubou thing, which is hilarious ID YouTube. Okay,

728
00:36:21,840 --> 00:36:25,000
so here's a pro tip for dot net developers. Okay,

729
00:36:25,039 --> 00:36:28,960
So think of coding agents as so agent mode chat.

730
00:36:29,199 --> 00:36:31,800
You know, code completions inside of VS, inside of VS

731
00:36:31,800 --> 00:36:36,360
code model context switch. Coding agents. These are that, but

732
00:36:36,400 --> 00:36:39,840
now they're working on some other machine doing work asynchronously

733
00:36:39,920 --> 00:36:43,760
for you. It's like having a whole plethora of coworkers

734
00:36:43,800 --> 00:36:46,960
assigning and doing work all at the same time, multiple branches,

735
00:36:47,000 --> 00:36:49,679
multiple things, things like that, and to get up Copilot

736
00:36:49,719 --> 00:36:52,480
coding agent is one of those same thing right now.

737
00:36:53,159 --> 00:36:56,440
Inherently what happens here, think of it is it needs

738
00:36:56,440 --> 00:36:59,320
to spin up architecture, needs to spin up a machine

739
00:37:00,079 --> 00:37:01,880
to write the code and run your code and test

740
00:37:01,920 --> 00:37:05,639
the code on it. So just like you would you

741
00:37:05,679 --> 00:37:08,840
write a GitHub action, or you would say, hey, I

742
00:37:09,239 --> 00:37:11,119
want this to be on a Windows VM, I wanted

743
00:37:11,159 --> 00:37:13,119
to have dot Net nine, I want to insult these workloads,

744
00:37:13,119 --> 00:37:15,440
blah blah blah blah blah, you gotta do the same

745
00:37:15,599 --> 00:37:18,920
thing for the coding agent. So you have to create

746
00:37:18,960 --> 00:37:21,800
a workflow, and you need to create it with a

747
00:37:21,800 --> 00:37:25,880
specific name, which is the Copilot setup instructions. And I

748
00:37:25,920 --> 00:37:28,199
had this happen. It was so upsetting everything that you're

749
00:37:28,239 --> 00:37:32,079
talking about in general. I'm opposed to link there. This

750
00:37:32,159 --> 00:37:34,679
is to mine. And what I do is I say, hey,

751
00:37:34,679 --> 00:37:36,519
you're going to run this on an am boontu Latest,

752
00:37:36,840 --> 00:37:39,239
give it read permission, check out the code, set up

753
00:37:39,280 --> 00:37:43,000
dot net nine, and then insull dependencies and insult the ASPIRECLI.

754
00:37:43,320 --> 00:37:46,000
And when you do that, what ends up happening is

755
00:37:46,039 --> 00:37:48,159
it writes the code and then it builds your project,

756
00:37:48,519 --> 00:37:51,360
so you can think of it like this. What happened

757
00:37:51,360 --> 00:37:54,920
to Carl is it tried to run the code and

758
00:37:54,960 --> 00:37:57,480
it's like I only got done at eight. I don't

759
00:37:57,519 --> 00:38:00,079
have dot at nine. So instead of it, does it

760
00:38:00,159 --> 00:38:02,639
necessarily know how to install stuff, you know on your

761
00:38:02,679 --> 00:38:06,199
behalf on that machine, you need to tell it.

762
00:38:05,320 --> 00:38:06,159
Speaker 1: It would be cool if it did.

763
00:38:06,360 --> 00:38:07,599
Speaker 4: To be honest with you, I mean it can you

764
00:38:07,639 --> 00:38:11,119
get restore, it can run those commands. So it's like, hey,

765
00:38:11,199 --> 00:38:13,199
I know, I know how to fix this, Like, oh,

766
00:38:13,199 --> 00:38:14,760
I see the problem. I see the problem.

767
00:38:14,800 --> 00:38:16,159
Speaker 1: I'll just eight.

768
00:38:16,280 --> 00:38:18,719
Speaker 4: I'm on a machine that's done at eight, but you

769
00:38:18,760 --> 00:38:21,280
want down nine, so I'm just going to downgrade it automatically.

770
00:38:21,480 --> 00:38:23,760
So that's how you get around that. And the same

771
00:38:23,800 --> 00:38:26,280
thing for for Maui, for example, if you're doing Mali

772
00:38:26,320 --> 00:38:30,000
work you install the Maui workload for for Android and

773
00:38:30,039 --> 00:38:31,280
I just run it on a boon joke because you

774
00:38:31,280 --> 00:38:33,159
don't need to inherently run it on iOS and all.

775
00:38:33,079 --> 00:38:33,639
Speaker 1: These other things.

776
00:38:33,800 --> 00:38:35,599
Speaker 4: So that's how that's how you set it up. It

777
00:38:35,639 --> 00:38:38,119
seems silly, but once on at ten's here, then the

778
00:38:38,360 --> 00:38:42,840
default right, the default machine, Well, then get upgraded. So

779
00:38:43,480 --> 00:38:44,960
That is a pro tip because I ran to that

780
00:38:45,039 --> 00:38:48,559
same thing. I use coding agents all the time. I

781
00:38:48,559 --> 00:38:50,719
have an idea at midnight, I open up the GitHub

782
00:38:50,760 --> 00:38:52,920
app on my phone, you know, I create an issue.

783
00:38:53,119 --> 00:38:54,840
Speaker 1: Boom done. Yeah.

784
00:38:54,880 --> 00:38:57,239
Speaker 4: And now if I'm on a branch, I just you

785
00:38:57,360 --> 00:38:59,440
just go in and there's an agent's panel to say, yeah,

786
00:38:59,480 --> 00:39:01,920
this app. Go. I wake up in the morning, you're

787
00:39:02,000 --> 00:39:04,559
like five pool requests ready for your review, you know.

788
00:39:04,639 --> 00:39:07,280
And it's like this crazy thing. Once you go all

789
00:39:07,320 --> 00:39:09,840
in that you can really build in hip code if

790
00:39:09,880 --> 00:39:11,920
you get it into that state where I think you're

791
00:39:11,960 --> 00:39:13,360
in the flow kind of like when we talk about

792
00:39:13,360 --> 00:39:15,440
code flow, right, you're in the addit area blah blah

793
00:39:15,480 --> 00:39:15,880
blah blah.

794
00:39:15,960 --> 00:39:16,119
Speaker 1: Right.

795
00:39:16,280 --> 00:39:20,199
Speaker 4: If you can get in in the flow with these agents,

796
00:39:20,199 --> 00:39:21,880
and I think it can be super super productive.

797
00:39:22,119 --> 00:39:23,679
Speaker 2: So I know, we got to take a break here,

798
00:39:23,719 --> 00:39:25,159
so let's do that. When we come back, I have

799
00:39:25,159 --> 00:39:27,800
a message about Azure deevop, so we'll be right back

800
00:39:27,840 --> 00:39:30,920
after these very important messages. Did you know there's a

801
00:39:30,960 --> 00:39:35,480
dot net on AWS community. Follow the social media blogs,

802
00:39:35,519 --> 00:39:39,800
YouTube influencers and open source projects and add your own voice.

803
00:39:40,360 --> 00:39:43,559
Get plugged into the dot net on Aws Community at

804
00:39:43,599 --> 00:39:51,000
aws dot Amazon dot com, slash dot net, and we're back.

805
00:39:51,039 --> 00:39:54,519
It's dot net rocks. I'm Carl Franklin, that's Richard Campbell. Hey,

806
00:39:55,079 --> 00:39:58,639
that's James Montemagno. And we're talking about AI and visual

807
00:39:58,679 --> 00:40:03,039
studio code in other places. So, yeah, I have a

808
00:40:03,079 --> 00:40:06,800
customer where we started out in GitHub and then they

809
00:40:07,400 --> 00:40:11,760
said we had to go over to Azure DevOps because

810
00:40:11,960 --> 00:40:15,239
there are other stuff, their legacy stuff is over there whatever.

811
00:40:16,079 --> 00:40:21,199
And then the GitHub copilot agent coding agent comes out

812
00:40:21,199 --> 00:40:23,760
and I'm like, geez, I wish I had this over there,

813
00:40:23,800 --> 00:40:26,000
but I don't. And so now I'm trying to get

814
00:40:26,079 --> 00:40:28,559
him to come back over to GitHub right and they

815
00:40:28,639 --> 00:40:31,639
won't do that, and I'm just like, what.

816
00:40:33,599 --> 00:40:34,760
Speaker 1: Can we do about that?

817
00:40:34,760 --> 00:40:38,880
Speaker 4: That's a great question. So there are a few integrations

818
00:40:38,960 --> 00:40:41,960
with get up copilot for azur DevOps. One, there's been

819
00:40:42,400 --> 00:40:45,800
GitHub Advanced Security for a while, which is pretty cool,

820
00:40:46,760 --> 00:40:50,159
so you can turn that onto scanning. There's also now

821
00:40:50,320 --> 00:40:53,960
Azure boards integration for get hub copilot. This is a

822
00:40:54,039 --> 00:40:56,000
private preview you can kind of see where this is going.

823
00:40:56,760 --> 00:41:01,400
But basically how this will work is that if you

824
00:41:01,480 --> 00:41:05,360
have using Azure DevOps and if your code is hosted

825
00:41:05,400 --> 00:41:09,360
on GitHub. Because Azure DevOps can connect to multiple Git repositories,

826
00:41:10,000 --> 00:41:14,519
you can then assign work through Azure DevOps to getthub

827
00:41:14,519 --> 00:41:16,719
code pilot and will perform that work for you on

828
00:41:16,760 --> 00:41:19,440
your back right. So it is a hybrid flow today.

829
00:41:19,920 --> 00:41:21,440
But that condition infrastructure.

830
00:41:22,079 --> 00:41:25,079
Speaker 2: That condition means that the code repositories have to be

831
00:41:25,079 --> 00:41:27,599
on GitHub and that's not where they are on TFS,

832
00:41:27,639 --> 00:41:29,039
and that's whe're going to stay there.

833
00:41:29,199 --> 00:41:31,280
Speaker 1: Which really what I thought was why how as your

834
00:41:31,280 --> 00:41:33,880
DevOps are supposed to work? That was the TFS approach,

835
00:41:34,199 --> 00:41:36,760
where GitHub actions was the GitHub appro.

836
00:41:36,719 --> 00:41:39,719
Speaker 4: Yeah, you know, I think that there's been a lot

837
00:41:39,760 --> 00:41:44,239
of listening and learning to Obviously, you know, the companies

838
00:41:44,320 --> 00:41:46,280
using Azure DevOps, we use a lot of Azure DevOps

839
00:41:46,280 --> 00:41:48,360
internally as well, so this is a problem for us

840
00:41:48,400 --> 00:41:50,800
to inherently we have tons of code on get up,

841
00:41:50,840 --> 00:41:53,079
but also tons of code not on GitHub. And there's

842
00:41:53,079 --> 00:41:55,320
a lot of teams that are like exactly in carl

843
00:41:55,360 --> 00:41:56,639
Spot that want to do this. So there's a lot

844
00:41:56,679 --> 00:41:58,880
of work being done there. I don't have necessarily like

845
00:41:59,280 --> 00:42:02,000
insight into it, but you know, the teams are listening

846
00:42:02,119 --> 00:42:06,440
and you're starting to see some of that listening back

847
00:42:06,480 --> 00:42:09,000
into product right away. Right inherently, there are two very

848
00:42:09,000 --> 00:42:11,840
different products that work very very differently. And the thing

849
00:42:11,960 --> 00:42:13,960
is do you build the thing twice?

850
00:42:14,320 --> 00:42:14,480
Speaker 1: Right?

851
00:42:14,599 --> 00:42:15,280
Speaker 4: You know what I mean?

852
00:42:15,880 --> 00:42:17,280
Speaker 1: Or that's what they suggested.

853
00:42:17,679 --> 00:42:21,400
Speaker 2: Actually they well they suggested I clone the repo in

854
00:42:21,480 --> 00:42:24,519
the kidthub and then use all the coding agent stuff

855
00:42:24,519 --> 00:42:27,280
and then move it back into the TFS repunt. I'm like, God,

856
00:42:27,280 --> 00:42:29,320
I want to do that that.

857
00:42:29,599 --> 00:42:35,119
Speaker 4: There's a few things though, I will say this is

858
00:42:35,119 --> 00:42:37,559
is that that's a little bit tricky in general on

859
00:42:37,639 --> 00:42:39,480
that ideal. Obviously for all the stuff that you're doing

860
00:42:39,480 --> 00:42:42,480
in VS and VS code, obviously all works just fine,

861
00:42:42,639 --> 00:42:47,199
just that coding agent part that is there for asynchronous work.

862
00:42:47,360 --> 00:42:51,119
And the thing is also like remote indexing. I don't

863
00:42:51,119 --> 00:42:53,960
think we have remote inducing for Azure DevOps, but if

864
00:42:53,960 --> 00:42:56,320
your code is on GitHub, when you're working inside of

865
00:42:56,400 --> 00:43:00,360
VS code or Visual Studio, we actually remote index your repo.

866
00:43:00,480 --> 00:43:02,760
And what that means is it can do semantic search

867
00:43:02,800 --> 00:43:05,320
a whole lot faster. And that means that all of

868
00:43:05,360 --> 00:43:08,800
your agent mode requests are like way faster, especially on

869
00:43:09,000 --> 00:43:11,960
huge codebases. And that's also super important. So I'm not

870
00:43:11,960 --> 00:43:13,599
sure that's going to come to Azure devlops for what

871
00:43:13,639 --> 00:43:15,639
that will look like, that would be pretty cool. I'm

872
00:43:15,639 --> 00:43:17,679
sure there's a future requests out there, but that is

873
00:43:17,719 --> 00:43:20,400
something to think about. But yeah, you know, I think

874
00:43:20,440 --> 00:43:22,880
with this world of AI, there's a lot of implications

875
00:43:22,920 --> 00:43:26,000
of where the tools are, what tools are you using,

876
00:43:26,039 --> 00:43:28,079
how do you blend these together? And also the team.

877
00:43:28,159 --> 00:43:30,440
But one thing we'll point out is, you know, I

878
00:43:30,519 --> 00:43:32,159
work with a lot of companies that they're not only

879
00:43:32,239 --> 00:43:35,280
using dot net spoiler alert. You know, they're not only shocking,

880
00:43:35,440 --> 00:43:36,440
they're not only.

881
00:43:36,239 --> 00:43:36,920
Speaker 1: Using vs code.

882
00:43:37,000 --> 00:43:39,880
Speaker 4: Right, they're in Exco, they're in Intelligent, they're in other ideas.

883
00:43:39,880 --> 00:43:42,400
You have a team that's maybe building mobile apps natively, well,

884
00:43:42,400 --> 00:43:45,079
they're in exco, they're inside of Android Studio. Maybe they're

885
00:43:45,079 --> 00:43:48,119
back ends in dot net. So do you buy five

886
00:43:48,159 --> 00:43:49,840
different AI coding projects?

887
00:43:49,920 --> 00:43:51,880
Speaker 1: Yeah, well, I'm seeing that in a lot of organizations

888
00:43:51,920 --> 00:43:55,280
where they're quite fragmented. There's guys running Windsurfing, guys running Cursor,

889
00:43:55,400 --> 00:43:57,719
and yep, they're all working against the same codebase. But

890
00:43:58,559 --> 00:44:01,920
some weird things happen sometimes, and the models right inherently

891
00:44:01,920 --> 00:44:04,519
are the same models across all of them right in general,

892
00:44:04,559 --> 00:44:07,039
just different user experience. But I will point out this

893
00:44:07,519 --> 00:44:12,519
get up Copilot does have extensions for excode for Intelligent

894
00:44:12,679 --> 00:44:15,039
for Eclipse. Right, that's out there, so you can actually

895
00:44:15,159 --> 00:44:17,760
use you know, Claude so on it, you know, powered

896
00:44:17,760 --> 00:44:20,199
by get up Copilot inside of xcode to write swift

897
00:44:20,239 --> 00:44:23,280
code like that exists today. That's out there. And I

898
00:44:23,320 --> 00:44:28,000
will say, if you're being enterprise, get up Copilot and

899
00:44:28,039 --> 00:44:29,960
the enterprise and business skeus give.

900
00:44:29,800 --> 00:44:30,800
Speaker 4: You a lot of.

901
00:44:32,280 --> 00:44:32,719
Speaker 1: Control.

902
00:44:32,760 --> 00:44:37,000
Speaker 4: For example, uh, my wife, we were just doing some

903
00:44:37,039 --> 00:44:39,000
coding and I was showing her all these models and

904
00:44:39,039 --> 00:44:40,360
this and this that, and then she's like, oky, oh

905
00:44:40,400 --> 00:44:42,079
I got we're doing like a little She goan to

906
00:44:42,119 --> 00:44:44,239
go do this thing. She's like, I only see GPT

907
00:44:44,320 --> 00:44:46,039
four oh and four one. She's like, I want sown it.

908
00:44:46,079 --> 00:44:47,760
Where's my son it? Where's my GPT five? And I

909
00:44:47,760 --> 00:44:49,800
was like, hell, your IT department needs to turn it

910
00:44:49,840 --> 00:44:51,679
on because it's like for her company. Right, So it's like, yeah,

911
00:44:51,840 --> 00:44:54,400
there's that control. So I think that's one thing too,

912
00:44:54,480 --> 00:44:55,440
is what do you want?

913
00:44:55,519 --> 00:44:57,000
Speaker 1: Right, I've asked. I've also seen you.

914
00:44:57,159 --> 00:45:02,199
Speaker 4: Unfortunately some folks that just are like contractors. Right, they

915
00:45:02,239 --> 00:45:04,320
work for a bunch of different companies, and then they

916
00:45:04,360 --> 00:45:07,239
get their get up account onboarded by the company temporarily,

917
00:45:07,800 --> 00:45:11,559
and then their get up copilot settings get overruled by

918
00:45:11,599 --> 00:45:14,000
the company processes. Right, So how do you manage that?

919
00:45:14,679 --> 00:45:18,599
It's a very interesting thing about, you know, different policies

920
00:45:18,639 --> 00:45:21,679
that are out there, so good or bad, right, how

921
00:45:21,719 --> 00:45:24,559
things are working? But yeah, it's not probably about there's CLIs,

922
00:45:24,559 --> 00:45:27,239
there's coding agents, there's integrations, there's all sorts of things

923
00:45:27,239 --> 00:45:31,199
and many tools. Obviously I use our stuff that we build.

924
00:45:31,320 --> 00:45:32,159
Speaker 1: I'm a little.

925
00:45:31,960 --> 00:45:34,880
Speaker 4: Biased, obviously, but I think our stuff's really great, and

926
00:45:35,280 --> 00:45:38,599
I've been really building and shipping a lot of code.

927
00:45:38,639 --> 00:45:41,320
But kind of to Carl's point earlier, it's also hard

928
00:45:41,360 --> 00:45:43,800
to really keep up with all this stuff, right, You.

929
00:45:43,800 --> 00:45:45,960
Speaker 1: Can't keep hopping between tools all the time. You got

930
00:45:45,960 --> 00:45:48,760
to get some work done too, exactly right. And so

931
00:45:48,840 --> 00:45:51,280
one thing that has been really nice is some into

932
00:45:51,280 --> 00:45:54,960
the newer standardization when it comes to things like MCP servers,

933
00:45:55,039 --> 00:45:57,800
model contacts and protocol servers that want to work everywhere.

934
00:45:57,920 --> 00:46:00,760
It isn't the embracing of thems CP just to proof

935
00:46:00,800 --> 00:46:04,119
that the industry is desperate for some standards, maybe a

936
00:46:04,159 --> 00:46:06,880
little bit because MCP is not great, but at least

937
00:46:06,880 --> 00:46:08,920
it's something people could agree. Yeah, it's a way to

938
00:46:08,920 --> 00:46:12,559
provide that additional context to data to these models, right,

939
00:46:12,639 --> 00:46:16,119
And you know there's tons of folks across the industry

940
00:46:16,159 --> 00:46:19,559
that are on the board, including folks from Microsoft and

941
00:46:19,599 --> 00:46:22,159
GitHub and the registry as well. So standardization of how

942
00:46:22,199 --> 00:46:24,159
does this work? How does it work in businesses? But

943
00:46:24,199 --> 00:46:27,119
also agents dot MD. You go to agent dot MD

944
00:46:27,280 --> 00:46:29,199
that is a kind of sid in the show link

945
00:46:29,280 --> 00:46:32,039
already because it's well worth a look and everybody seems

946
00:46:32,039 --> 00:46:33,559
to be conceding that too. Yeah.

947
00:46:33,639 --> 00:46:36,800
Speaker 4: You know, here's the problem as well, is like how

948
00:46:36,800 --> 00:46:39,360
do you make sure that you have unique features versus

949
00:46:39,400 --> 00:46:42,239
everyone else? So it takes time to create those standards,

950
00:46:42,280 --> 00:46:45,159
but creating time means like this thing was created like

951
00:46:45,199 --> 00:46:47,280
a few months ago. Now it's standard, which are kind

952
00:46:47,280 --> 00:46:49,159
of crazy to think about in this modern day. So

953
00:46:49,239 --> 00:46:53,039
Agents at MD is basically a way of open format

954
00:46:53,079 --> 00:46:55,400
for coding agents that works across all those things that

955
00:46:55,440 --> 00:46:58,320
you just mentioned, including vs code. The coding agent will

956
00:46:58,360 --> 00:47:02,199
come to Visual Studios soon obviously, and CLIs and it

957
00:47:02,280 --> 00:47:05,480
is exactly like copilot instructions in a.

958
00:47:05,440 --> 00:47:07,880
Speaker 1: Way, it's like a universal language.

959
00:47:07,960 --> 00:47:10,360
Speaker 4: It's a universal language. So what we see is folks

960
00:47:10,400 --> 00:47:13,920
will mix and match these agents and copilot instructions have

961
00:47:13,960 --> 00:47:17,760
really specific things what they're using a get up copilot,

962
00:47:17,760 --> 00:47:19,760
and then more generic things for the agent's ot MD

963
00:47:19,920 --> 00:47:22,199
that anything can work on. But yeah, so it's great

964
00:47:22,199 --> 00:47:24,000
to see so many names there, and I think you'll

965
00:47:24,000 --> 00:47:27,079
see more of this sort of standardization as it goes.

966
00:47:27,119 --> 00:47:29,719
But also tools will do unique things and they'll stand

967
00:47:29,760 --> 00:47:30,320
out as well.

968
00:47:30,480 --> 00:47:32,639
Speaker 1: So yeah, yeah, I got to push back on the

969
00:47:32,719 --> 00:47:36,599
vibe coding term. It kills me. I mean because what

970
00:47:37,280 --> 00:47:40,320
Caparthy was talking about and it was only earlier this year,

971
00:47:40,360 --> 00:47:44,599
which is crazy to think about that. It really was

972
00:47:45,039 --> 00:47:48,840
like senior developers should experiment, but you know, you don't

973
00:47:48,840 --> 00:47:52,280
deploy those experiments where I feel like you're using it more.

974
00:47:52,320 --> 00:47:55,239
You're using your tools as in a PM role, And

975
00:47:55,280 --> 00:47:57,159
you really say vibe coding and we say, I'm not

976
00:47:57,199 --> 00:48:00,000
writing the code, but I am supervising the process.

977
00:48:00,000 --> 00:48:03,840
Speaker 2: It's like a pejorative that got reclaimed by another culture, right.

978
00:48:04,000 --> 00:48:08,000
You know, so my kids used to call me bougie.

979
00:48:08,800 --> 00:48:12,000
You know, Oh that's so that's so boogie dad. You know,

980
00:48:12,119 --> 00:48:15,119
when I'm like cooking a steak that costs sixty bucks

981
00:48:15,159 --> 00:48:18,199
or something like that, Oh, it's so bougie. And then

982
00:48:18,679 --> 00:48:21,320
you know, I'm like, I'm going to use that term

983
00:48:21,360 --> 00:48:23,599
in a positive way. I'm gonna, Oh, this is gonna

984
00:48:23,639 --> 00:48:28,039
be the boogiest dinner you've ever had, right, And they're like, no, Dad, No,

985
00:48:28,840 --> 00:48:29,519
you can't do that.

986
00:48:31,559 --> 00:48:35,599
Speaker 4: I think if I'm going the same thing, it's hard.

987
00:48:35,679 --> 00:48:37,559
When I think of vibe coding is really just coding

988
00:48:37,559 --> 00:48:39,239
with AI. Let's just be honest with you and you're

989
00:48:39,280 --> 00:48:42,119
in a flow, just like you can be coding anytime.

990
00:48:42,400 --> 00:48:44,360
To me, honestly, that's all vibe coding is. I went

991
00:48:44,400 --> 00:48:46,960
on hansom In's podcast and he had the same exact thing,

992
00:48:47,239 --> 00:48:49,000
I hate you know, blah blah blah. I was like, listen,

993
00:48:49,599 --> 00:48:53,239
I think the term is silly, but it also describes

994
00:48:54,199 --> 00:48:56,199
what is happening in my mind. I was up with

995
00:48:56,280 --> 00:48:58,920
David Fowler working on the feedback flow app because him

996
00:48:58,960 --> 00:49:01,239
and I were going back and forth on it, and

997
00:49:01,840 --> 00:49:03,280
it's like this is a real production now, like this

998
00:49:03,320 --> 00:49:05,000
is a thing that actually ships, like a real money

999
00:49:05,000 --> 00:49:08,079
for my personal subscription that has It's one of the

1000
00:49:08,119 --> 00:49:10,199
biggest things that I've ever shipped, you know, in the

1001
00:49:10,280 --> 00:49:13,079
last fourteen years since I did advocacy. It's it's a

1002
00:49:13,079 --> 00:49:15,679
real product. And I also vibe code tons of tiny

1003
00:49:15,679 --> 00:49:18,719
projects as well. But it is just really coding. But

1004
00:49:18,800 --> 00:49:20,159
I am in a vibe like I'd be up at

1005
00:49:20,159 --> 00:49:22,360
two in the morning going back and forth and David's like,

1006
00:49:22,400 --> 00:49:24,639
what if you had this new feature and literally five

1007
00:49:24,639 --> 00:49:26,719
minutes later, I'd have it pushed to production. It was

1008
00:49:26,960 --> 00:49:31,079
like the vibes, the flow. It's a coding flow, just

1009
00:49:31,079 --> 00:49:32,920
with AI. So the vibe coding part.

1010
00:49:32,760 --> 00:49:34,079
Speaker 1: Is it's all vibes.

1011
00:49:34,119 --> 00:49:36,280
Speaker 4: I say, it's vibes all the way down. I turn

1012
00:49:36,360 --> 00:49:39,639
on some music. I'm just going. And I told my wife,

1013
00:49:39,639 --> 00:49:42,480
I said, when I was really deep in this project,

1014
00:49:42,639 --> 00:49:44,960
really in the beginning, I said, I'm sorry, I'm just

1015
00:49:45,039 --> 00:49:48,519
I haven't had this much fun coding in like a decade,

1016
00:49:48,920 --> 00:49:50,760
and I can't help myself. And I know I'm gonna

1017
00:49:50,840 --> 00:49:55,320
wake up refresh and I'm going I managing a project.

1018
00:49:55,360 --> 00:49:58,159
I'm not not coding. I am supervising in real time.

1019
00:49:58,199 --> 00:50:00,360
I don't know the code is mine. I'm prompt, I'm

1020
00:50:00,400 --> 00:50:03,760
doing things. I got into this flow where literally as

1021
00:50:03,800 --> 00:50:07,360
it was churning and it was describing what's doing and it's

1022
00:50:07,360 --> 00:50:12,119
doing I'm actively reviewing the code so fast. In my mind,

1023
00:50:12,239 --> 00:50:15,000
that flow, that flow state was just there. I'm not

1024
00:50:15,079 --> 00:50:16,519
say it's going to be there for everyone when you're

1025
00:50:16,519 --> 00:50:18,480
working on very pointed updates, but it happened.

1026
00:50:18,599 --> 00:50:20,679
Speaker 1: No, No, And I totally and I agree with what

1027
00:50:20,719 --> 00:50:22,960
you're saying too, Like, I think this is a great point.

1028
00:50:23,039 --> 00:50:25,800
It's like, if your passion is typing code, you're not

1029
00:50:25,840 --> 00:50:27,880
going to love these tools. Now if your passion is

1030
00:50:27,880 --> 00:50:32,239
delivering solutions to customers, boy, this thing cranks out solutions fast.

1031
00:50:32,320 --> 00:50:35,760
And you're running in there that PM role of supervising

1032
00:50:35,800 --> 00:50:39,519
the overall flow, but you're also responsible for the architecture.

1033
00:50:39,800 --> 00:50:41,800
You know, you're you're kind of put on the QA head. Ever,

1034
00:50:41,840 --> 00:50:43,559
so I woulday, are we really going down the right path?

1035
00:50:43,559 --> 00:50:46,199
Like you're pressing on a lot of things. But and

1036
00:50:46,239 --> 00:50:49,239
that could be really fun. I have a friend who said, dude,

1037
00:50:49,320 --> 00:50:52,599
I've learned how to fly, and I'm really enjoying flying.

1038
00:50:53,039 --> 00:50:58,320
Speaker 4: I worked Okay, so not ALLAI coding is vibes right,

1039
00:50:58,360 --> 00:51:01,280
So I worked with co pilot pother and I implemented

1040
00:51:01,440 --> 00:51:05,039
authentication into my feedback flow app. It took one month

1041
00:51:05,480 --> 00:51:09,480
one poll request. I added twenty we added twenty two

1042
00:51:09,519 --> 00:51:11,800
thousand lines of code, removed five thousand lines of code.

1043
00:51:12,000 --> 00:51:15,159
We had two hundred files changed in a conversation of

1044
00:51:15,199 --> 00:51:17,880
two hundred and forty comments back and forth. We're having

1045
00:51:17,960 --> 00:51:21,280
a conversation reviewing code back and forth in real time, right,

1046
00:51:21,559 --> 00:51:24,159
And this happened pushing back and forth to get us right.

1047
00:51:24,199 --> 00:51:26,719
Speaker 1: And to be clear, you're conversing with software.

1048
00:51:26,840 --> 00:51:29,880
Speaker 4: I am conversing with software as if it is. It

1049
00:51:29,920 --> 00:51:32,480
is Carl sitting next to me, and and I'm reviewing

1050
00:51:32,480 --> 00:51:34,280
code with Carl. That's what I'm.

1051
00:51:34,079 --> 00:51:36,679
Speaker 2: Doing, James.

1052
00:51:36,840 --> 00:51:39,280
Speaker 1: But I also have a great appreciation for putting all

1053
00:51:39,360 --> 00:51:43,039
that as issues and in pull requests. Commentary gives you

1054
00:51:43,039 --> 00:51:46,480
a clear documentation of what the intent was and what

1055
00:51:46,519 --> 00:51:47,639
the tools did for you.

1056
00:51:47,880 --> 00:51:51,039
Speaker 4: Yeah, one hundred percent, one one hundred percent and there,

1057
00:51:51,119 --> 00:51:54,599
like I am, the guidance is just a tool, whether

1058
00:51:54,880 --> 00:51:58,159
whether it's there or not, just like intellcode or Intelecentre. Honestly,

1059
00:51:58,199 --> 00:51:59,079
it's exactly the same.

1060
00:51:59,119 --> 00:52:02,719
Speaker 1: And I'm high resisted the personalization because people are overdoing it, right,

1061
00:52:02,760 --> 00:52:05,239
They're taking this to the r but do I talk

1062
00:52:05,280 --> 00:52:08,000
to my car and my car misbehaves? Damn, right, I do.

1063
00:52:09,199 --> 00:52:09,760
There you go.

1064
00:52:10,440 --> 00:52:12,760
Speaker 4: I will say this. I do talk to agents and

1065
00:52:12,840 --> 00:52:14,760
AI much different than I would talk to Carl A

1066
00:52:14,920 --> 00:52:17,119
hundred percent sure, are.

1067
00:52:16,679 --> 00:52:18,840
Speaker 1: Per I know a lot of folks just like if

1068
00:52:18,880 --> 00:52:21,320
you conversed with a person the way you're talking to

1069
00:52:21,360 --> 00:52:24,679
this tool, HR would be calling you. Right, this is

1070
00:52:24,760 --> 00:52:26,000
abusive language.

1071
00:52:26,159 --> 00:52:28,039
Speaker 2: You know what I talk about when I talk about

1072
00:52:28,039 --> 00:52:30,880
the Amazon thing that starts she who starts with a right.

1073
00:52:31,159 --> 00:52:33,960
We can't really say her name because she does cause harm.

1074
00:52:34,159 --> 00:52:37,440
So my wife, when you know, I will usually say,

1075
00:52:37,800 --> 00:52:41,360
you know, hmmm, stop when she's rambling on about something

1076
00:52:41,360 --> 00:52:43,880
I wanting to stop. My wife says shut up, and

1077
00:52:43,920 --> 00:52:46,440
it works, but I feel bad.

1078
00:52:47,639 --> 00:52:51,360
Speaker 1: Yeah, you know like negative language, negative language and it works.

1079
00:52:51,360 --> 00:52:52,360
It affects both ways.

1080
00:52:52,559 --> 00:52:55,159
Speaker 4: Yeah, I think just looking at it is like if

1081
00:52:55,159 --> 00:52:57,159
I'm writing the code or I'm writing the prompt. It's

1082
00:52:57,199 --> 00:53:00,119
like I was going to write this code eventually, right,

1083
00:53:00,320 --> 00:53:02,519
but I'm doing the code reviews. I'm like you said,

1084
00:53:02,519 --> 00:53:03,599
I'm in charge of the archestras.

1085
00:53:03,880 --> 00:53:05,039
Speaker 1: You are responsible.

1086
00:53:05,280 --> 00:53:07,079
Speaker 4: Yeah, And I would say this even if you're like,

1087
00:53:07,079 --> 00:53:08,840
I don't necessarily wanted to write my code. I was

1088
00:53:08,880 --> 00:53:11,880
sitting down with a developer the other day and this

1089
00:53:12,079 --> 00:53:15,280
really complex application, big scale application, and they were hitting

1090
00:53:15,280 --> 00:53:17,840
this endpoint. It was making a call off to EF

1091
00:53:17,840 --> 00:53:19,639
and hitting their database and it was just returning a

1092
00:53:19,639 --> 00:53:22,039
five hundred er no exception. It just was five hundred

1093
00:53:22,119 --> 00:53:25,039
coming back in the swagger. And I said, Okay, here's

1094
00:53:25,039 --> 00:53:26,639
what I wreck And they were like, I've been trying

1095
00:53:26,639 --> 00:53:28,039
to figure out what the heck this is going for

1096
00:53:28,039 --> 00:53:30,599
the last forty five minutes. I said, take the entire exception,

1097
00:53:31,079 --> 00:53:34,360
go into ask mode and put on a thinking model, right,

1098
00:53:34,400 --> 00:53:36,719
a thinking model that can do deep research. Say hey,

1099
00:53:36,920 --> 00:53:39,199
I have this method in this thing. When I call

1100
00:53:39,320 --> 00:53:42,440
this this is happening. I know the breakpoint goes here,

1101
00:53:42,519 --> 00:53:43,760
it doesn't hit the next line.

1102
00:53:43,920 --> 00:53:44,719
Speaker 1: What's the problem.

1103
00:53:44,920 --> 00:53:47,760
Speaker 4: Within thirty seconds, it's like, Oh, what actually happened is

1104
00:53:47,760 --> 00:53:50,039
you'd set up a new model and the DTO and

1105
00:53:50,159 --> 00:53:53,119
enerny framework is expecting it as a foreign key, but

1106
00:53:53,199 --> 00:53:56,920
it expects it as a different name inherently, so you

1107
00:53:57,000 --> 00:53:58,719
need to add an attribute to foreign key.

1108
00:53:58,840 --> 00:53:59,079
Speaker 1: Right.

1109
00:53:59,239 --> 00:54:00,400
Speaker 4: Could they have figured that out?

1110
00:54:00,639 --> 00:54:01,840
Speaker 1: Yes? How long would it took?

1111
00:54:01,920 --> 00:54:02,519
Speaker 4: Who knows?

1112
00:54:02,719 --> 00:54:04,280
Speaker 1: But it's smart. Could have been the whole day and

1113
00:54:04,320 --> 00:54:07,760
your keyboard imprints on your forehead right like being there thrashing.

1114
00:54:08,000 --> 00:54:11,079
Speaker 4: So I think it is it is a do you

1115
00:54:11,159 --> 00:54:13,119
need to go all in to this agent mode and

1116
00:54:13,159 --> 00:54:16,159
it's coding agent like start slow, right, get some suggestions,

1117
00:54:16,320 --> 00:54:19,400
ask it, get some insight into the application. I was

1118
00:54:19,440 --> 00:54:23,960
working with the with copilot just yesterday with my MCP

1119
00:54:24,079 --> 00:54:26,159
server I have for this app, and I was like, yeah,

1120
00:54:26,320 --> 00:54:28,320
you know, how do I architect it? What does this

1121
00:54:28,400 --> 00:54:30,480
look like? What do I need to update my application?

1122
00:54:30,599 --> 00:54:33,360
It came together with a plan. It's like, here's the plan.

1123
00:54:33,440 --> 00:54:36,119
Let's review the plan, and then let's implement the plan. Right,

1124
00:54:36,559 --> 00:54:39,559
So I'm in control, right human in the loop, I'm

1125
00:54:39,599 --> 00:54:43,079
always in control of everything that it's doing. I'm prompting,

1126
00:54:43,159 --> 00:54:45,559
I'm creating the plan, I'm creating the issues. I'm reviewing

1127
00:54:45,559 --> 00:54:48,719
that code. I'm the one pressing the merge button right right,

1128
00:54:48,800 --> 00:54:50,760
I'm merging the code right at the end of the

1129
00:54:50,840 --> 00:54:54,599
day that I've reviewed inside of it. And the hope

1130
00:54:55,199 --> 00:54:57,840
is that with the guidance of these instructions and these

1131
00:54:57,880 --> 00:55:00,679
agents on MD that that the code that it generates

1132
00:55:00,920 --> 00:55:03,199
is very similar to code that I would generate at

1133
00:55:03,199 --> 00:55:04,679
the end of the day, so I have to review

1134
00:55:04,719 --> 00:55:07,960
and the compiler always gets to say and the requirement

1135
00:55:08,000 --> 00:55:09,960
stock is sitting there to compare against.

1136
00:55:10,039 --> 00:55:12,480
Speaker 1: Like I do think we're in a unique place in

1137
00:55:12,599 --> 00:55:18,320
our role as creators of software to take these generative

1138
00:55:18,320 --> 00:55:21,280
AI tools to places that are harder for other industries

1139
00:55:21,280 --> 00:55:23,199
to do, just because we're used to vetting this and

1140
00:55:23,239 --> 00:55:26,639
we're used to building goal posts and evaluating against them.

1141
00:55:26,679 --> 00:55:30,039
Like we have a lot of tools and behaviors long

1142
00:55:30,119 --> 00:55:34,199
before the LM showed up that support this development practice.

1143
00:55:34,760 --> 00:55:36,719
Speaker 4: It's a mind shift. It's just a mind shift, just

1144
00:55:36,760 --> 00:55:39,400
like any tool or extension that you've ever in soaved

1145
00:55:39,400 --> 00:55:42,239
individual study or VISKA. It is how do I learn

1146
00:55:42,360 --> 00:55:45,400
this tool? How do I become an expert at this

1147
00:55:45,719 --> 00:55:48,960
tool to help me be more productive? As little or

1148
00:55:49,000 --> 00:55:51,639
as much as you want to use you know, it depends.

1149
00:55:51,719 --> 00:55:51,880
Speaker 1: You know.

1150
00:55:51,960 --> 00:55:54,440
Speaker 4: It's just a slider bar, that's all it is. And

1151
00:55:54,519 --> 00:55:57,800
to me, it's am I saving a minute, an hour

1152
00:55:58,159 --> 00:56:01,480
a day? All of those are important. It's just a

1153
00:56:01,679 --> 00:56:03,400
I can shave a few minutes, that's great.

1154
00:56:03,679 --> 00:56:05,719
Speaker 2: It's another choice that you take before you sit down

1155
00:56:05,719 --> 00:56:07,320
and write a line of code. It's like all right,

1156
00:56:07,400 --> 00:56:11,880
here's here's what I have to do. HM, can sign it?

1157
00:56:11,960 --> 00:56:12,719
Help me out here?

1158
00:56:12,880 --> 00:56:13,039
Speaker 4: Yeah?

1159
00:56:13,159 --> 00:56:14,840
Speaker 1: Can you know is.

1160
00:56:14,760 --> 00:56:16,599
Speaker 2: It going to be faster for me to do it

1161
00:56:16,639 --> 00:56:18,760
myself because I know, or is it going to be

1162
00:56:18,800 --> 00:56:21,480
faster for me to create a prompt and blah blah

1163
00:56:21,519 --> 00:56:24,400
blah and explain things and have the AI do it.

1164
00:56:24,519 --> 00:56:28,079
Speaker 4: Example, if I want to like rename files or move

1165
00:56:28,119 --> 00:56:31,159
things around or actually like change in name space, like

1166
00:56:31,440 --> 00:56:34,440
Visual Studio is very good at that and it's very proficient. Right,

1167
00:56:34,480 --> 00:56:39,360
if you move a folder from models to models, person

1168
00:56:39,480 --> 00:56:41,880
people or whatever, you need to update the name spaces,

1169
00:56:41,920 --> 00:56:44,800
It'll just do it for you. You could have prompted the LM.

1170
00:56:45,079 --> 00:56:47,639
Now you're just burning tokens for no reason. Let the tools.

1171
00:56:47,719 --> 00:56:50,519
Let the tools be awesome at what tools are doing,

1172
00:56:50,639 --> 00:56:53,559
right vs Code and Visual Studio are awesome at so

1173
00:56:53,639 --> 00:56:56,760
many things without AI right that it's doing, you know,

1174
00:56:56,960 --> 00:57:00,039
just inherently the static analysis. Let it be awesome at that,

1175
00:57:00,360 --> 00:57:02,760
and then use the model for these harder problems to

1176
00:57:02,840 --> 00:57:03,840
kind of solve as well.

1177
00:57:03,719 --> 00:57:05,199
Speaker 1: Or just nuisance things.

1178
00:57:05,320 --> 00:57:07,719
Speaker 2: Right, So I had I don't know a list of

1179
00:57:07,840 --> 00:57:11,119
twenty properties that I had to boolly and check and

1180
00:57:11,199 --> 00:57:14,199
if if the properties were true, then I had to

1181
00:57:14,320 --> 00:57:17,440
express some HTML, right, and I had to do that

1182
00:57:17,480 --> 00:57:19,840
for like twenty properties, but they weren't in alphabetical order,

1183
00:57:19,920 --> 00:57:21,920
so you'd end up with a list with you know,

1184
00:57:22,000 --> 00:57:24,199
names of things all over the place. And I just

1185
00:57:24,360 --> 00:57:26,480
you know, told the I think it was chat Chiptia,

1186
00:57:26,599 --> 00:57:28,440
just said can you alphatize this for me?

1187
00:57:28,519 --> 00:57:29,599
Speaker 1: And yeah, no problem.

1188
00:57:29,920 --> 00:57:33,719
Speaker 4: Copy paste the vs code website. We have these dev

1189
00:57:33,760 --> 00:57:36,639
days going on, these community events. All the data is

1190
00:57:36,639 --> 00:57:38,840
in like a CSV file. So I go into agent

1191
00:57:38,880 --> 00:57:40,280
MO and I say, hey, I need to add And

1192
00:57:40,320 --> 00:57:42,559
then inside the vs code websites a bunch of like

1193
00:57:42,599 --> 00:57:45,760
you know, different blobs, different data, different things, and I say, hey,

1194
00:57:45,840 --> 00:57:48,360
I need to add these five events. Just go update

1195
00:57:48,400 --> 00:57:50,800
the site with these five events, and go update in

1196
00:57:50,840 --> 00:57:53,159
these places and it'll take. It's like, this is CSV data.

1197
00:57:53,239 --> 00:57:55,599
It knows how to transform that and just put it.

1198
00:57:55,800 --> 00:57:58,320
And then we also have locations of latitude and longitude.

1199
00:57:58,360 --> 00:58:00,880
So I'm like, and also go figure out the latitude

1200
00:58:00,920 --> 00:58:06,079
and longitude. It knows the latitude and longitude of you know, Portland, Oregon,

1201
00:58:06,199 --> 00:58:08,519
for example, Right, I can figure it out, and of

1202
00:58:08,519 --> 00:58:10,400
course I review it to make sure it's correct. It's

1203
00:58:10,400 --> 00:58:12,639
not one hundred percent every time. And that's another part

1204
00:58:12,719 --> 00:58:14,800
in part right it, it'd be clear it doesn't know

1205
00:58:15,079 --> 00:58:16,519
it looked it up. It looked it up, and it

1206
00:58:16,559 --> 00:58:17,639
may have looked it up incorrectly.

1207
00:58:17,800 --> 00:58:19,119
Speaker 1: That's let Yeah, I like it.

1208
00:58:19,559 --> 00:58:23,480
Speaker 4: It does it exactly exactly. It's all the context that

1209
00:58:23,559 --> 00:58:24,000
it has.

1210
00:58:24,119 --> 00:58:25,880
Speaker 2: They all have brain damage. You got to pick your

1211
00:58:25,880 --> 00:58:28,119
brain damage basically.

1212
00:58:27,599 --> 00:58:30,559
Speaker 1: Well, let's start with they don't have brain Oh, come on, Richard,

1213
00:58:30,679 --> 00:58:34,360
lighten up. It's just software. It is just soft software.

1214
00:58:34,679 --> 00:58:37,840
Speaker 2: But it's easier for humans to engage with something when

1215
00:58:37,840 --> 00:58:40,519
they treat it like another human. That's what we do

1216
00:58:40,599 --> 00:58:43,920
with our dogs and our pets and our cars and everything.

1217
00:58:44,199 --> 00:58:47,280
Speaker 4: I think it's a it's a big database. I think

1218
00:58:47,280 --> 00:58:49,800
it was a big database that it's quarrying, right, But

1219
00:58:49,960 --> 00:58:53,119
what it does is that it can querry that database

1220
00:58:53,239 --> 00:58:54,679
kind of like you're saying, Richard's like, it's just a

1221
00:58:54,679 --> 00:58:56,440
big database of stuff that it's pulling from.

1222
00:58:56,599 --> 00:58:59,000
Speaker 1: But what it does is it.

1223
00:58:58,199 --> 00:59:00,239
Speaker 4: Summarizes things and a thing that I can part Like

1224
00:59:00,280 --> 00:59:03,119
I was just looking at example example out of the blue.

1225
00:59:03,599 --> 00:59:05,840
We our dog has diabetes and we have to give

1226
00:59:05,880 --> 00:59:09,000
her shots every day, the VET accidentally gave us you

1227
00:59:09,079 --> 00:59:11,639
one hundred needles. We're supposed to get you forty needles.

1228
00:59:11,679 --> 00:59:13,440
I actually have no idea what that means. Are they

1229
00:59:13,519 --> 00:59:16,079
the same? The units look okay, they're different, Like they

1230
00:59:16,119 --> 00:59:18,239
are the same. So I just went into copilot on

1231
00:59:18,320 --> 00:59:21,000
my at Microsoft copilot and I said, just like you're

1232
00:59:21,039 --> 00:59:23,079
going shat Gypt, I said, what's the difference between you

1233
00:59:23,199 --> 00:59:25,719
forty and you one hundred needles? And give me an

1234
00:59:25,760 --> 00:59:28,079
analysis of like what this means and if there's conversions

1235
00:59:28,159 --> 00:59:30,800
or anything. And it gave me tables, It gave me graphs,

1236
00:59:30,519 --> 00:59:32,719
and it gave me warnings like these are not the same.

1237
00:59:32,760 --> 00:59:34,920
You should not be using these internal No, it's two

1238
00:59:34,920 --> 00:59:37,159
and a half times more. Now I could do all

1239
00:59:37,199 --> 00:59:39,079
of that research, all of that research. All that data

1240
00:59:39,119 --> 00:59:40,639
exists on the internet, and if I went to Google

1241
00:59:40,719 --> 00:59:42,719
or went to bing right, they give me that. But

1242
00:59:42,719 --> 00:59:44,800
it gave me this nice thing that I could then share,

1243
00:59:44,880 --> 00:59:46,760
I could keep in my memory. I now understand that I

1244
00:59:46,760 --> 00:59:48,880
could go back to that's all these things are big

1245
00:59:48,960 --> 00:59:52,280
databases of information, and it knows how to nicely summarize

1246
00:59:52,280 --> 00:59:55,119
things for us in human form in a nice way

1247
00:59:55,159 --> 00:59:58,440
that I can understand or tell it, to tell it

1248
00:59:58,480 --> 00:59:59,719
to display it in a different way.

1249
00:59:59,800 --> 01:00:02,599
Speaker 2: Right, here's a creative Here's an issue though, Right, A

1250
01:00:02,639 --> 01:00:05,719
lot of people are using these AI things to get

1251
01:00:05,800 --> 01:00:09,039
to gather facts and then they're satisfied with the fact

1252
01:00:09,039 --> 01:00:14,159
that the facts look okay, and so you know, we

1253
01:00:13,559 --> 01:00:16,800
we we're not going to double check because that's why

1254
01:00:16,840 --> 01:00:19,079
we asked it in the first place, because we didn't

1255
01:00:19,119 --> 01:00:21,039
want to go to Google and go to different resource

1256
01:00:21,079 --> 01:00:23,800
places and look things up. So I think that people

1257
01:00:23,840 --> 01:00:26,519
are doing that, they're just you know, asking for facts,

1258
01:00:26,559 --> 01:00:30,360
taking the facts as they are, not checking them and

1259
01:00:31,280 --> 01:00:34,000
using them. I don't think that's right.

1260
01:00:33,960 --> 01:00:35,719
Speaker 4: Now, I think no, I agree. I think it's the

1261
01:00:35,719 --> 01:00:37,480
same thing with coding, right, like the code that's in

1262
01:00:37,519 --> 01:00:39,960
the flow, Like I'm reviewing, I'm understanding, I'm doing that.

1263
01:00:40,320 --> 01:00:43,320
I I I'm not when I say I'm you know,

1264
01:00:44,079 --> 01:00:46,320
just in the flow and I'm just like accepting the stuff.

1265
01:00:46,360 --> 01:00:49,320
I'm reviewing the stuff I'm reviewing what's generating, right, I'm

1266
01:00:49,360 --> 01:00:50,519
not having an auto commit.

1267
01:00:50,760 --> 01:00:50,920
Speaker 1: Right.

1268
01:00:50,960 --> 01:00:53,000
Speaker 2: But in this case, you didn't know the difference between

1269
01:00:53,039 --> 01:00:56,320
you forty and U two hundred, right, so you couldn't know.

1270
01:00:56,840 --> 01:00:58,199
Speaker 1: Could not It was accurate or not.

1271
01:00:58,320 --> 01:01:01,719
Speaker 4: Now that being sad, like you said, it's up. It's

1272
01:01:01,960 --> 01:01:05,280
just like it's just like funnily enough going to googler

1273
01:01:05,320 --> 01:01:07,760
bang and then opening a bunch of links. How much

1274
01:01:07,760 --> 01:01:09,760
do you trust those sources as well to it?

1275
01:01:09,880 --> 01:01:10,119
Speaker 1: Right?

1276
01:01:10,320 --> 01:01:12,719
Speaker 4: So, the one thing that I do like about and

1277
01:01:12,760 --> 01:01:16,599
what I how I use AI. Actually, surprisingly I use

1278
01:01:16,760 --> 01:01:20,159
very little like AI chapbot, I don't use JGBT. I

1279
01:01:20,239 --> 01:01:22,159
use Copilot a little bit for some research. It's not

1280
01:01:22,199 --> 01:01:24,199
really my I haven't gotten into that flow of like

1281
01:01:24,480 --> 01:01:27,360
how I want to quarry the Internet for this data,

1282
01:01:27,880 --> 01:01:30,920
But I will say that the one thing that I

1283
01:01:31,039 --> 01:01:32,920
do and how I use it is it gives me

1284
01:01:32,960 --> 01:01:34,480
the links of where I got the research.

1285
01:01:34,840 --> 01:01:36,159
Speaker 1: I always leble check. But I agree with you.

1286
01:01:36,239 --> 01:01:38,079
Speaker 4: I don't think people are doing that, And I think,

1287
01:01:38,320 --> 01:01:40,760
how do you bubble that up? So people know at

1288
01:01:40,840 --> 01:01:42,880
least at high level did this come from? You know,

1289
01:01:43,039 --> 01:01:45,039
CNN dot com? Did this come from the.

1290
01:01:45,039 --> 01:01:47,119
Speaker 2: That I want at least three sources?

1291
01:01:47,360 --> 01:01:47,559
Speaker 1: Right?

1292
01:01:47,840 --> 01:01:50,920
Speaker 2: And if if any one of those three sources is different,

1293
01:01:51,440 --> 01:01:52,400
now we have a problem.

1294
01:01:52,639 --> 01:01:55,239
Speaker 4: Yeah, and you can always ask it to double check itself.

1295
01:01:55,360 --> 01:01:58,159
I had it come up with some numbers. We're celebrating

1296
01:01:58,159 --> 01:02:00,760
some number like fifty million Visual City users. It's like,

1297
01:02:00,760 --> 01:02:02,679
give me some fun stats about fifty million. It just

1298
01:02:02,719 --> 01:02:04,320
came up with something and was like, yeah, but really

1299
01:02:04,360 --> 01:02:06,280
is that really true? And it's like, yeah, some of

1300
01:02:06,280 --> 01:02:10,480
them weren't. I was like, okay, you know, ask it,

1301
01:02:10,599 --> 01:02:13,400
like are you sure that that is writing? Is that correct?

1302
01:02:13,519 --> 01:02:14,920
It's like, well, I was just trying to be fun.

1303
01:02:14,880 --> 01:02:16,800
Speaker 1: Like okay, well, I was asking the crack type.

1304
01:02:16,960 --> 01:02:19,000
Speaker 4: I was like being serious here, like let's go in

1305
01:02:19,079 --> 01:02:20,960
you know, on this. So I think it is like

1306
01:02:21,000 --> 01:02:26,360
with anything, yeah, any new tool, you know, making sure

1307
01:02:26,360 --> 01:02:29,760
that you are reviewing all of all the input and

1308
01:02:29,800 --> 01:02:32,079
output that that you have. Right, you can give it

1309
01:02:32,079 --> 01:02:33,719
bad input and get bad output. You can give a

1310
01:02:33,719 --> 01:02:35,360
good input and get bad output. You can get good

1311
01:02:35,360 --> 01:02:38,280
input and get good output. But always having that human

1312
01:02:38,320 --> 01:02:41,440
in the loop is super important and reviewing everything in

1313
01:02:41,480 --> 01:02:45,280
the world. Right, but also the same Right, if I

1314
01:02:45,320 --> 01:02:46,920
walk down the road and I talked to five people

1315
01:02:46,960 --> 01:02:49,440
and asked them about something, I might get five different answers, right,

1316
01:02:49,599 --> 01:02:51,519
which one's correct? I don't know what are their sources.

1317
01:02:51,880 --> 01:02:53,159
So that's how you look at it.

1318
01:02:53,159 --> 01:02:55,639
Speaker 2: It makes a good case for us as developers because

1319
01:02:55,880 --> 01:02:58,079
we are the guys with the knowledge and we can

1320
01:02:58,159 --> 01:03:00,719
see whether that code that it general it is right

1321
01:03:00,840 --> 01:03:06,559
or wrong, or efficient or inefficient. So it really really

1322
01:03:06,639 --> 01:03:10,079
really works. AI really works for us. It's just in

1323
01:03:10,159 --> 01:03:15,440
that whole fact based MELIU that I think problems arise

1324
01:03:15,480 --> 01:03:16,280
and then it'll get better.

1325
01:03:16,360 --> 01:03:19,280
Speaker 1: Well, and if you know the code is generating is

1326
01:03:19,320 --> 01:03:21,559
in theory a set of facts about executing. It's just

1327
01:03:21,599 --> 01:03:24,400
that there's an easy way to validate it. Yeah. But

1328
01:03:24,719 --> 01:03:28,320
I think this behavior of you know, building a validation

1329
01:03:28,440 --> 01:03:31,599
strategy out of every output is just going to be

1330
01:03:31,679 --> 01:03:33,400
part of the reflex the same way.

1331
01:03:33,559 --> 01:03:35,239
Speaker 2: Yeah, And I just don't think many people are going

1332
01:03:35,320 --> 01:03:35,679
to do it.

1333
01:03:35,760 --> 01:03:38,519
Speaker 1: Well, they will eventually. We all got better at searching

1334
01:03:38,559 --> 01:03:41,079
the internet too. Sure, right like the early days it

1335
01:03:41,199 --> 01:03:43,960
was worse. It's one would argue the peak has come

1336
01:03:44,000 --> 01:03:47,760
and gone and it's worse. It's getting worse again. You know,

1337
01:03:47,800 --> 01:03:50,840
we're back to this mechanism of having to validate more

1338
01:03:50,920 --> 01:03:54,239
often and more completely, and for multiple sources. You know,

1339
01:03:54,280 --> 01:03:56,639
the behaviors will change over time.

1340
01:03:57,039 --> 01:04:00,960
Speaker 2: Well, this has been a fascinating dive into more AI,

1341
01:04:01,079 --> 01:04:03,880
and especially with Visual Studio code and Visual Studio James.

1342
01:04:03,880 --> 01:04:06,119
Thank you very much. It's always a pleasure to doc

1343
01:04:06,159 --> 01:04:06,920
to you, my friend.

1344
01:04:06,920 --> 01:04:10,320
Speaker 1: Thanks for having me entirely too long since the last time. Yeah, anytime.

1345
01:04:10,320 --> 01:04:12,039
Speaker 4: I'm sure if you have me back in two months,

1346
01:04:12,079 --> 01:04:12,880
everything will be completely.

1347
01:04:12,920 --> 01:04:16,320
Speaker 1: Everything will be difference, all right.

1348
01:04:16,360 --> 01:04:18,760
Speaker 2: And check out coded with AI dot com and we'll

1349
01:04:18,760 --> 01:04:20,400
talk to you next time on.

1350
01:04:20,440 --> 01:04:21,159
Speaker 1: Dot net rocks.

1351
01:04:41,840 --> 01:04:44,559
Speaker 2: Dot net Rocks is brought to you by Franklin's Net

1352
01:04:44,679 --> 01:04:48,639
and produced by Pop Studios, a full service audio, video

1353
01:04:48,719 --> 01:04:52,840
and post production facility located physically in New London, Connecticut,

1354
01:04:53,039 --> 01:04:57,840
and of course in the cloud online at pwop dot com.

1355
01:04:58,039 --> 01:05:00,440
Visit our website at d O t any t r

1356
01:05:00,480 --> 01:05:04,440
O c k s dot com for RSS feeds, downloads,

1357
01:05:04,559 --> 01:05:08,239
mobile apps, comments, and access to the full archives going

1358
01:05:08,280 --> 01:05:11,519
back to show number one, recorded in September two.

1359
01:05:11,360 --> 01:05:11,920
Speaker 1: Thousand and two.

1360
01:05:12,599 --> 01:05:14,920
Speaker 2: And make sure you check out our sponsors. They keep

1361
01:05:15,000 --> 01:05:18,159
us in business. Now go write some code, see you

1362
01:05:18,199 --> 01:05:21,280
next time. Got middle Vans

1363
01:05:23,440 --> 01:05:23,480
Speaker 1: And