One of the biggest blocker for innovators is timely and fit for purpose information. This is a cool product. Beats scheduled tasks in Gemini or chatgpt for me given how good the Scira AI Exa combo is. ++ cheaper to run than @ManusAI_HQ

If only it wasn't a subtitle typo in Netflix and ACP Pradhyum was actually asking Chatgpt for clues in the case 🤣🤣🤣 what a closed loop it would be to CID 🤣 IFYKYK

Introducing the new Rork Agent for mobile apps ・90% less errors with error auto-fixes ・New powerful agentic tools ・10x smarter with best practices from Claude Code ・Produces the best UI – using both GPT-5 & Sonnet 4 The best agent for mobile apps is here 👇

We just raised a $9.4M seed round to seize the means of production of software. Turns out, it fits in your pocket. vibecode (Thread)

I am so loving the generate image tooling in @lovable_dev! For some very content heavy pages it does a magnificent job!

Got some extra placeholder invites.. Who wants them?

Hang on.. is OpenAI Developers Codex gpt-5-codex high running on ~1M token context length?????

Over the past week, Arc Institute published three new discoveries that I’m very proud of. • The world's first functional AI-generated genomes. Using Evo 2 (the largest biology ML model ever trained, which Arc released in partnership with NVIDIA in February), Arc scientists

Extrapolation of ARCs amazing foundational work in non human domains esp Agriculture and environmental applications will truly unlock long term value, even beyond human health where data is available in more excess

Today, my boyfriend William Fedus launched his company, Periodic Labs. I feel so lucky to have witnessed it from day one—Liam is the hardest-working person I know and the most passionate about science. From the start, our home became a gathering place for some of the brightest

This is a great step and one of the only actual efforts I believe will solve a long term biosecurity concern with generated proteins/sequences. A lot of AI Biosecurity work assumes only viruses and bacteria can do bad - which is grossly naive. The next step is to now inform

Can Claude team PLEASEE take a look at the classifiers on sonnet 4.5 and opus 4.1? Whatever you guys did last week has made it useless for biology research in Claude.ai if i even breathe the word biosec..... or virus ... even flagged the biosecurity patch Science

For all those who test agentic scaffolds or frameworks daily- which have you found works the best for scientific research? I'm refreshing & tossing between LangChain deep agents, Claude agent sdk, CrewAI . But am I missing anything? Want to boost co-scientist style

Today, we are launching Red Queen Bio (redqueen.bio), an AI biosecurity company, with a $15M seed led by OpenAI. Biorisk grows exponentially with AI capabilities. Our mission is to scale biological defenses at the same rate. A 🧵 on who we are + what we do!👇🏻1/19

Now this is a company type i like! Having spent years tracking biosecurity risks across the globe I strongly believe this is a critical necessity at all data flow junctures so we can be proactive. A vast majority of biosecurity responses are delayed because someone has to look.

SO MANY GEMINIs.. I hope they are good at science