Skip to main content

Jay Parikh

Nov 6, 2026

2:23
Can you talk to us a little bit about where your head is at? the way that we think about building products and where the traditional software engineer fits in that future of how we build software?
2:39
So one, absolutely think things are changing and they're evolving quickly.
2:45
And I would say that the world in that is focused on building products also is a very dynamic range today, right? Because you have enterprises that are still working one way and then you have maybe certain, you know, maybe it's just a one person company that's working a completely different way, right? And so we see a very big range of how developers and builders or product engineers want to work today.
3:18
And in Core AI, but really in GitHub, we have to make sure that the platform, the surfaces, the offerings really do encompass or offer enough to everybody to move forward and get that work done.
3:34
Now, for us, I would also say that it's important to think through the outcomes, right? And in some ways, we have to be more and more focused, and we can because we don't have to spend as much time typing lines of code by hand and doing different things by hand.
3:52
Now, if you are savvy with these technologies, a lot of that, I would say, toil work, you know, being able to go and do this type of data analysis or gather these things or go integrate these various things that were manual and maybe kind of clunky.
4:10
These are things that you're like, hey, I just don't want to have to do this.
5:34
Where in your mind is the line between what a software engineer does or a product engineer and a product manager? Is there a line or are these kind of becoming the same thing?
2:22
So I mean, maybe building on that, how do you define excellence in engineering leadership?
2:27
Yeah, so for me, there's like five things maybe that come together.
2:30
One is you have to build a high-performing team, so that is all about how do you recruit the top people, you know.
2:38
And then thinking about, like, the composition of the team, right? Because you need to think about do you want...
2:44
How many early career folks do you want? How many specialists do you want? How many generalists do you want? How many senior tech lead type of people do you want? Senior ICs, et cetera.
2:54
Second then is the, in some ways, like, setting that strategy, setting that vision.
2:59
Like, where is that North Star going? But it's gotta be at the right level, because if it's too high level, then it's too opaque.

52 MINS LATER

55:10
Exactly
15:37
It is, yeah, and hearing about some of the, you know, agentic harnesses that build their own skills from it, and I think you announced some of those today, how do you test reliability and correctness of, like, an agent that builds its own skills?
15:51
So I-- One is it comes back to just the importance of the evals, right? Really grounding in as much time, effort you can put into having evals that objectively tell you what's right, what's wrong, and that isn't a once and done thing.
16:07
As those agents continue to operate in the enterprise, you want those evals to get better and better and better, right? So you wanna shape those.
16:15
The second thing then is we have things in, in Foundry, for example, where you can set up those guardrails where you can actually send a red team agent to test your agent to see where it may go off the paths that you expect, and it'll grade you.
16:32
It'll give you the suggestions for where you need to improve, you know, tool calling or your context, your prompt or your skills or all of the above.
16:42
So we'll guide you in the platform with that red team testing in addition to the evals, which are those things that you develop that they should be running, not just at deploy time, but they should be continually running, right? So this is an area where I think as we work with customers, and I would say, you know, listen, Ryan, a very iterative process.
17:03
I would not say it's done at this point, but again, we're thinking about this at the platform, right? So you should be able to bring your own evals.

7 MINS LATER

23:53
What sort of thinking do you put into the, the harness and the models?
48:10
And you're describing what's happened since.
48:12
So, you know, we should start with the fact that beginning, since the beginning of the year, the amount of traffic that has come to GitHub has gone up significantly, right? So last year we had about a billion commits that we processed in all of 2025.
48:26
Now we process about 300 million commits each, each weekSo traffic has gone up pretty significantly, and we're very committed to improve the availability, the performance of the platform.
48:40
We're getting help from more of the Microsoft engineers and the top talent that we have.
48:45
We're attracting new talent to the team.
48:47
We're bringing in more capacity to scale, and we're re-architecting these systems.
48:52
But, you know, I feel really good about the path that we're on.
49:09
This is right
speaker_0HOST
91:00
... and really makes, makes more sense to just focus on what you have?
91:02
Yeah, I think there's, like, an element to, you know, developers, builders, uh, around craft.
91:07
And I think you're always gonna want to find, like, the best tool or the tools that suit your sense of craft, right? Whether you're a w- wood ...
91:17
Like a wood maker or you're a painter.
91:20
You know, and, and there's like a, a big element of craft, so I think that there's gonna be use cases where, hey, this is the way to do it.
91:26
These are the best tools to, say, modernize or upgrade some version of, like, old Java code that you may have.
91:33
And there may be just, like, this is the one or two just true tried ways of doing it and proper tools that you use.
speaker_1HOST
93:56
Yeah.

We value your privacy

We use cookies to understand how you use our platform and to improve your experience. Click “Accept All” to consent, or “Decline non-essential” to opt out of non-essential cookies. Read our Privacy Policy.