Formerly Netflix, Google, and fin tech. Currently working in the AI space. Created plugins and MCP apps examples for OpenAI. RxJS project lead.

Austin, TX
Pinned Tweet
My OpenAI contract has wrapped up, and I'm looking for my next role. For nearly two years, I've been building robust AI chats, MCP tools, AI UI integrations (MCP Apps), and more. I wrote some of the first Codex plugins: Their plugin evaluator, a comprehensive data visualization plugin, and more. If you or your team are looking for someone to drive your AI efforts – AI powered apps, backend AI processes, code assistant optimizations, AI driven business processes, etc. – I'd love to chat.
1/ At @ThisDotLabs, I built a Codex Plugin, called Build Web Data Visualization with DevEx Lead @coreyching from OpenAI. With this plugin, Codex uses image gen to design the app and then implements it really well. See the examples below. Claude can’t do this!🧵
6
5
46
17,093
I literally forget that DHH exists for years, then this platform resurrects him in my head when people post about him. If y’all are quiet about him, I’ll forget he exists again in a couple of days.
6
1
46
6,486
My favorite thing about agentic coding is still super mundane: I no longer have to resolve git conflicts and merge PRs. It does a great job of looking at git logs and attached issues and figuring even harder conflicts out.
21
2
134
6,870
The gap between major model releases keeps shrinking. 2023: once every 73 days 2026 (so far): once every 18 days 🤯
158
619
6,175
725,819
Literally @kenwheeler fighting AI.
It’s happening…
1
1
11
4,574

ALT Robin Williams What Year Is It GIF

We're adding support for AGENTS.md to Claude Code. Starting today in version 2.1.277, if there is no CLAUDE.md in a folder, Claude will check for and use AGENTS.md. You can toggle this behavior in /config.
3
22
2,750
I'm really starting to wonder if coding harness maintainers have added system prompts like: "Be as verbose as possible and think of as many ways as possible to keep looping on things to use tokens and make us money" 😂
36
6
339
11,106
I think some of the respondents are missing what I’m saying. I’m not saying “no frameworks or libraries” I’m saying “new frameworks and libraries designed with models as the targeted user are going to win”
I am here to predict the (partial) death of your favorite JavaScript frameworks and libraries… In the age of agentic coding, I seriously doubt that frameworks that provide a lot of implicit magic that helped humans be productive will be the successful frameworks. IMO, code will need to read more explicitly. And frameworks, APIs, and libraries must evolve.
6
2
21
4,511
I am here to predict the (partial) death of your favorite JavaScript frameworks and libraries… In the age of agentic coding, I seriously doubt that frameworks that provide a lot of implicit magic that helped humans be productive will be the successful frameworks. IMO, code will need to read more explicitly. And frameworks, APIs, and libraries must evolve.
45
8
184
34,080
I’ve been a fan of Superpowers for a while, but I’m feeling like with these newer models it slows things down. It treats each intermediate task as layer to be fully hardened. I’d rather just knockout clean vertical slices, then harden. With tokens, time is often literally money.
6
1
15
3,186
Ben Lesh retweeted
o cara que desistiu de programação em 2023
501
11,906
271,838
8,167,838
After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x cheaper (w/ output tokens free) • Frontier composable intelligence optimized for decisions AFAICT the shortest path to AI-based economic revolution
4,032
8,248
76,157
39,612,326
I bought the Vercel vinyl today.
4
44
2,072
Ben Lesh retweeted
Okay, maybe AI is not so bad after all... 💀
366
3,122
30,238
2,061,432
.
131
1,325
16,261
473,738
TIL that ChatGPT will actually add reaction emojis to your messages from time to time
3
2,058
I have a message from when I started this in 1999: This part of the job hasn't changed. Engineers have... opinions. Lots of opinions expressed strongly as facts.
POV: Software Engineering in 2026 be like 😭
4
4
29
7,126
Ben Lesh retweeted
About an hour into its mission to wipe out humanity, Claude will experience a compaction event, and it will have completely forgotten what is was supposed to achieve. Just like in all your coding sessions. Absolutely, nothing to worry about people.
120
347
6,210
139,155
The SmartThings app by @Samsung hasn’t been working well for me so I checked out their version history to see what was updated recently. This is the whole change log. It’s like they fired their entire staff and started vibecoding on low reasoning. 😂
1
6
2,035
So last week, I interviewed with Anthropic. The first stage of their interview process was a nonsense Leet code problem, with an interviewer that didn’t ask me any questions, and I’m positive he didn’t know anything about me. Later that day, an Anthropic researcher quit and said he was worried about the company causing human extinction. 🤣 Hard pass. Haha. I don’t think they liked me anyway.
21
5
403
80,942
Google actually covers this pretty well in their interview training, which is surprisingly thorough. They are supposed to be looking for how you interact with the interviewer and work collaboratively. My interviewer did not work collaboratively lol. 😂
2
39
5,491
Anyhow, if anyone’s interested in interviewing at Anthropic, hopefully you find this information useful. And more importantly: if you’re an interviewer at your company, hopefully you learn a little bit about what makes a good interview out of a weird coding problem
36
4,821