@cansar

Codex @OpenAI. Previously: Cash App & Block, YC S17, Google, several startups.

Joined August 2008
Pinned Tweet
A personal update: I’m joining OpenAI to work on Codex, the best agent app in the industry. After years building consumer products, my obsession since September has been to make Block one of the most AI-native engineering orgs in the world. I’m proud of what we’ve built: faster Eng velocity with RPI in October, an internal multi-harness ADE in January, Slack-based cloud agents in February, and AI woven into every step of product development. The team is just getting started. Along the way I tried almost every agent orchestrator out there — 30+ over the past year. Codex felt special from the day it launched. Lately, it’s started to point at something much bigger. What used to be scattered across prompts, tabs, tools, files, and handoffs is starting to come together in one place. The app is becoming the single surface for knowledge work — letting you accomplish more than you ever thought possible, while staying calm and focused. You can already see the early signs: making a slideshow and editing it right there, working inside a custom site built just for you and your team, handing off to the browser and the computer while you watch it go. Getting this right takes powerful models. But it also takes taste, product thinking, and real care for the person on the other end. You see that in the product, and you feel it the moment you talk to the team. I couldn’t be more excited to join the Codex team. Together we’re going to build the future of computing and knowledge work, and we’re going to build it for you.
90
17
5
739
118,096
I'm telling you. For the level that I'm interested in, the technical barriers are literally zero. I know this because I have zero technical knowledge. For awhile now, the only barriers has been gumption, and even there you need less and less of it.
2
4
1
71
10,904
Codex users: What kind of projects & repos do you usually work with? Large / Small? Established / New? Many / Few? Want to understand this more broadly.
8
2
10
2,229
Can Sar retweeted
A reminder that multiple things can be true at the same time. Things are rarely zero sum games.
13
12
1
217
6,918
I am continually puzzled with the CLI-as-a-UX fascination. Claude Code now emulates sidepane there - I mean, it is super-cool, but that's like being able to shave with a specially sharpened ax - impressive, but why? Just use/invest in the app, clearly a dominating strategy...
53
3
3
292
28,828
People are somehow still arguing about whether AI is or will be a big deal, huh?
2
11
865
I think philosophical commitment to longtermism as an ethical concept drove a lot of interest which led some EAs to AI safety as a concept, but it turns out literally nothing about AI policy hinges on the social discount rate. It's an intellectual history curiosity.
14
14
2
372
31,629
Can Sar retweeted
ok why are the prominent turkish americans in the us political / media ecosystem are so…weird? hasan piker, dr oz, cenk uygur…
8
1
36
4,758
I’m maybe biased now but I had this take before too. Don’t be reckless with tokens, but don’t overthink it either.
It's hard to justify using models that are dumber than Fable and Astra. For real-world code work, the benefits massively outweigh the cost. The benefit isn't better code, it's something else that's more subtle.
1
5
1,313
Can Sar retweeted
Starting today, you can connect multiple accounts with most plugins in ChatGPT! 🎉 Bring context from your work and personal accounts into the same conversation. Connect your accounts in the plugin directory: chatgpt.com/plugins Devs: this works automatically, with no changes needed. But, to make the experience for users of your plugin even better, add a profile tool to your MCP server so ChatGPT can label the accounts: developers.openai.com/plugin…
169
146
105
1,933
739,661
Can Sar retweeted
when using openai models, ADR (Architecture Decision Record)'s is all you need (tm) to keep an agent on track. you still need to engineer the backpressure from day 0 and continually revirew it but ADRs with openai models are the bees knees.
22
7
4
162
9,137
Housing is good.
4
462
Can Sar retweeted
We will fix this
181
43
63
2,067
226,541
Realized that I never upgraded to Tahoe on my personal machine because I've been too busy to use it the last year 🙃 Just installed it (need XCode for something) and saw the new Safari. It's certainly something...
3
8
1,558
Replying to @JainAI29 @grok
I don't code with the terminal anymore, that was last year's tech
38
32
24
607
202,891
Can Sar retweeted
today is a good day to note: I love America and I love its exceptional free speech norms, enshrined in law and protected by its people and culture. the ability to say true and unpleasant or destabilizing things is not a right most people (even in the Western world!) enjoy this extends beyond merely what the first amendment and related doctrine legally protects to the nurture and support of the entire American culture. this culture allowed for the creation of the free-spirited Internet forums that birthed so much of modern thinking on artificial intelligence. wherever are the sacred cows, the free internet is able to poke and prod them, for better and for worse the actual realized pdoom of our culture is much lower due to our ability to freely discuss things as off-putting and nasty as the chances of human extinction. even when powerful interests find this inconvenient, the discourse does not stop happening -in public, in excruciating detail, without needing anyone's blessing. companies allow or encourage this to a surprising degree, at least in part because they know the cultural immune reaction will be strong if surprising levels of free speech norms are not upheld, as seen in recent history I know many of you are sick of “discourse”, having bathed in its sludge, its many pathologies, its propagandists bigots and sycophants, but the planet would be in much worse shape if all of this technological development was happening in cultures or countries where free discourse was stifled. this norm survives only because all of us keep choosing to encourage it and enforce it even when it's costly
114
196
50
2,864
207,188
ChatGPT Image 2.5 has unlocked my dormant illustrator/art director energy. All the illustrations I've ever wanted, now within reach, (almost) exactly as I want them, in my style. I am going ham making illustrations for every slide in this talk. It's visual thinker Christmas.
23
20
3
839
33,102
One of the greats
Today is my last day at @CashApp. A very bittersweet moment. I’ve truly loved the past ~6 years here. Such a great team and deeply positive impact on so many customers’ lives. I now have a deeper respect for @jack than I did on day one and believe the future of @blocks can and will be amazing. I’ve learned so much from, and will deeply miss, the people here. Will share what’s next for me in a couple of weeks. 💚
3
19
2,548
ChatGPT casually watching and transcribing my conference talk practice run, identifying when I'm doing meta-commentary in the middle like "cut this slide" "update my script here", time stamping and matching it to my speaker notes, making edits in Keynote with computer use. Mad.
14
5
1
250
10,171
It's hard to overstate how dangerous speeding towards RSI is. That's why I + 1385 others signed the pacing the frontier petition asking the US government to pace AI development. I'm guesstimating this is ~8-10% of all frontier lab employees.
I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.
76
81
15
622
57,002