Rishabh Srivastava

I genuinely think we built the best search engine for official economic data. Been working on this for 6 months. We spent ~$100k in tokens to structure economic data and make it easier to search. It's answers economic data really well. From "What has been the actual impact of AI on software engineering jobs in the last 2 years?" to "Why did egg prices increase so much more than chicken prices in the last 5 years?" Would love feedback (the more blunt the better). We have a generous free tier for the next week!

@Yuchenj_UW Oh wow. All the best for your next thing!

I have decided to step down as CTO at Hyperbolic. Leaving a company you co-founded and poured your heart into is not easy. So many moments still feel vivid: launching our AI inference product for open-source models and seeing tens of thousands of developers sign up in a week; the week we were hit by a massive DDoS attack and the entire engineering team fought around the clock until we won; the day we launched the GPU platform and watched ARR take off. There were also hard moments. That’s the nature of building a startup. I’m grateful for all of it. What I’m most grateful for is the team. Thank you for your trust. Most startups never build something people want. I believe we did. You should be proud of yourselves. I will look forward to seeing your success. What’s next for me? I’m still figuring it out. I believe this is the most extraordinary moment in human history. We’re standing at the edge of the Singularity. AI will reshape everything, and I still feel the same excitement I felt when I first fell in love with AI. Time to start over. Time to climb another mountain. Thank you to everyone who has been part of the journey, — Yuchen

@waltertayannlee i use a refurbished m4 mac mini (bought for S$720). mostly cause I need the extra processing power and RAM for data-heavy workflows but any machine that you don't mind having to reset (because all it has is work backed up on git) is great IMO!

@rishdotblog Which sandbox do you recommend btw?

Sigh does everyone saying "it will be expensive to maintain AI generated code" not know about --dangerously-skip-permissions Give Claude Code / Codex a sandbox, let it poll for updates/canges, auto push and review PRs, set up CI/CD. Automated, better-than-human maintenance 🤷🏽‍♂️

sigh. i hope anthropic changes their auth provider soon (or better yet - use a homebrewed solution). their current one seems very broken

TIL while testing out a new feature. Damn. These sanctions bit pretty hard

@huybery Thanks for the amazing work you and the team have done for the open source community. Excited to see what you do next!

bye qwen, me too.

Vercel deployments failing because of a kinetic war was not on my bingo card

Worked from a rock next to the sea today The world is still what it is, but my head is a lot better. 10/10 would recommend going out and touching grass if your brain is in a weird place

@mayurekbote Partially! Started a couple of years before Covid though

@rishdotblog Glory days of labour were indeed a historical anomaly - but only due free money followed by excessive money printing due to Covid. It has little to do with technology

Reminder re Citrini report - the glory days of labor (2018-present) are a historical anomaly

@NirantK Was quite surprised to see this! Our generation somehow graduated into the best job market since 2000, and I think we all have weird expectations because of that 😅

@rishdotblog woah! this is a killer narrative graph

- hallucinating function names that don't existing during agentic workflows - hallucinating incorrect structures when asked to generate structured outputs sonnet 4.5 still works great, but 4.6 is completely crapping the bed on the same tasks

Did anthropic break something after releasing sonnet 4.6? seeing a ton of hallucinations eveywhere for both Opus 4.6 and Sonnet 4.6 cc: @trq212, @alexalbert__

Whatever this was seems fixed!

@swairshah ~0 on maintenance! our workflows were fairly standard and covered by test cases same for features. we didn't need polish (just needed the work done). so the initial version is still standing. the only maintenance we've done is updating api keys / model names

@rishdotblog Any take on maintenance? like is this $100/mo tool you replaced with your custom one how much time did you have to spend on fixes/features etc?

Lol this was 1 year ago (Sonnet 3.7). Capabilities improved much faster than expected. Diffusion slower than expected (though accelerating)

@NirantK aye fair enough!

@rishdotblog sir 90% of what humans *around me* do today is non-verifiable tasks what metrics should go in dashboard vs how to make dashboard are very different questions and take 30-60 min of interviewing minimum!

Reasonable to assume that I'm somewhat AI "native": I've written a NLP book for devs, trained a Hindi LM in 2019. I can promise you that humans mogg AI right now on quality/finish/polish. Sadly, it is very difficult to tell quality for anything outside my narrow expertise