Jonas Templestein
@jonas
CEO http://iterate.com, prev. cofounder/CTO Monzo, dad of two
2025 will be the year we see the first self-driving startups. Level 0: No AI People do everything. They come up with ideas, build products, and run operations. Many legacy businesses still work this way. Level 1: People use AI tools ⬅︎ we are here People might use ChatGPT…
Yes
maybe the way to handle offloading tasks to semi-reliable agents is to just rebuild everything around git. email? git. shopping? git. paying taxes? git. calendar? believe it or not, also git.
Building on durable objects is so fun. Took me 10 LOC to make our cloudflare agents subclass automatically resume LLM requests that got cancelled because the durable object crashed or got evicted or whatever
Does parallel tool calling not work in the openai responses API when streaming or something? Even when I set parallel_tool_calls: true, I can't get any model to start more than one function tool call per response
5 y/o had friends over and did a spy meeting. I found this afterwards: Plan to escape. Shoot the grownups. ❤️
Today, we at @OpenAI achieved a milestone that many considered years away: gold medal-level performance on the 2025 IMO with a general reasoning LLM—under the same time limits as humans, without tools. As remarkable as that sounds, it’s even more significant than the headline 🧵
1/N I’m excited to share that our latest @OpenAI experimental reasoning LLM has achieved a longstanding grand challenge in AI: gold medal-level performance on the world’s most prestigious math competition—the International Math Olympiad (IMO).
Looking for 1-2 AI maximalist typescript hackers to join an internal hackathon in London in person next Thursday evening to play with the first version of our operating system for software 3.0
Gemini refused to work on my source code file because there were variables named childState in it ?!?!
Massive accomplishment. Congrats @antonosika and team 🫡
Lovable just raised $200M at a $1.8B valuation led by Accel. This all started unexpectedly with me calling my friend at 6AM to go for a walk. I've never shared this story before: (thread)
Photoshop... but WITHIN Slack? 🔄 - Generate images, edit images - Work collaboratively with colleagues
Let’s savour this one last time
Humanity has prevailed (for now!) I'm completely exhausted. I figured, I had 10h of sleep in the last 3 days and I'm barely alive. I'll post more about the contest when I get some rest. (To be clear, those are provisional results, but my lead should be big enough)
Getting to that age where most people I meet don't even know who Guybrush Threepwood is
Programming with cursor and biome = my macbook gets so hot I can fry an egg on it and the battery lasts only one hour Not sure if skill issue
As slow and inaccurate as it will ever be
holy shit groq + k2 is fast af - video not sped up stuff is changing real fast
Imagine a Silicon Valley startup doing this. No over-strategizing. Just ship. I guess it’s not obvious this is actually a good strategy for Kimi, but it is certainly good for us!
We've heard your feedback — Kimi K2 is SLOOOOOOOOOOOOW 😭 Especially for agentic apps, output tokens per second really matters. The main issue is the flooding traffic and huge size of the model, we are actively working on inference optimization and BUY MORE MACHINES! Speed…
Cursor keeps suggesting irrelevant things like "User prefers the padding on the right side of the stream to be p-4 when working in this repo". Making a good memory system for LLMs is hard!