Anthropic Engineer on the Future of Coding with AI | Thariq Shihipar

**Thariq Shihipar** (0:00)
A lot of us on the team have stopped using Plan Mode because the model just inks correctly. The agents and this models overall are much more intelligent than maybe even right now we're utilizing them. If we were to freeze model development right now, we'd probably still be figuring out new capability overhang for another six or 12 months. Some people look at you when you're making a startup and be like, why are you doing this? It's not the safe thing to do anymore, and it was for a long time at Silicon Valley. You just have to be excited about the expiration and uncertainty.

**Evan Tana** (0:27)
Many of you maybe have met Thariq when he has come back and spent some time during our AI startup school demos which have been awesome, but it was so great that we wanted to give you a full hour just for conversation and Q&A. But as I mentioned, a member, now alum, and one of the faces of Claude Code.
But let's spend a little time bringing people up to speed on your journey, almost your minus one journey, and your time at Anthropic. You joined a year ago, and we haven't actually talked about this. What went into that decision? What brought you there? Then we'll get into a little bit of what the last year has been like, which I'm sure has been fascinating.

**Thariq Shihipar** (1:05)
Yeah, totally. I ran a VC-backed gaming company for five years before joining SPC.
I was trying to make it into an AI gaming company, and my co-founders would not do it. Around 2024, I decided to shut down the company and return funds, and I was debating between joining... I was talking to the cursor team a lot, because I'd actually been using cursor. But then I was like, no, I am going to try and start a new company, and I joined South Park Commons. And yeah, I think that I actually really felt that was a really pivotal moment for me, especially because of the small groups. So we were in a little squad, and in that squad, I met Tom, who now runs GoodFire, and Sawyer, who is one of the smartest AI engineers I know. Fire squad. Yeah, yeah, yeah. So I was just sort of going through it. I was sort of mourning my old company and trying to figure out like what, like could I make another company, or like was I good at this AI stuff or something, you know? And talking to Tom, who had like sort of led interpretability at DeepMind and sent me these papers, and I was like, oh yeah, like I saw that hard. You know, like I can start like, you know, doing work here and contributing. And so I worked for Tom's company for about three months. Yeah. And so after SPC, I worked for Tom's company. And then while I was doing that and wrapping that up, Sawyer texted me being like, have you used Claude Code?
One month. This was like March 2025 Yeah.
And he's like, Opus 4 just came out. It's so much better than Cursor. And I was like, no, I have it. But I got into it and I was immediately just like, really, really hooked. And I was like, yeah, I need to like sort of join Anthropic. And I like, doesn't matter what they want me to do there. I'll do it. You know, I think initially I joined as like, what do they call it? Like a demo designer on the design team. And I was just like, dude, like if you think that's what I can do, like, sure. You know what I mean? And then ultimately, like very quickly found myself on the Claude Code team, yeah.

**Evan Tana** (3:21)
I guess on that note, you have sort of become one of the faces of Claude Code. Was that intentional or again, I think you've historically liked technical writing, but like, how did that come about?

**Thariq Shihipar** (3:32)
Yeah. I mean, I think that my specific interest was sort of like trying to increase the communication bandwidth between humans and agents. And so I feel that the agents and these models overall are much more intelligent than maybe even right now we're utilizing them. When I joined the Claude Code team, one of the first features I introduced was this thing called Ask User Questions. So during the flow, especially during planning, Claude Code might be like, hey, let's stop and clarify what you're trying to do. And I think previous to this, people were just sort of like, it was very dependent on the user to in detail explain what they want and know the unknown unknowns, for example. And so that was something where I felt has become one of the most popular features in the Claude Code, really changed how people interact with the agent. But I also felt the more time I've spent on it, I actually feel like it's more important to educate humans than improve the agent UX sometimes.