Just onboarded our new employee, Reachy Mini is Founding Embodied AI Engineer @deepintxns

this is very much giving @bodleianlibs / antique book shop vibes

@OpenAI's first hacker house. ever. 250k up for grabs. london, june 15–21. ifykyk #basedhouse

Speed and quality get the wow. But the dark horse behind @deepintxns has been our AI designer agent, powered by OpenAI's GPT Image 2 model. Sneak peek: Agent designed customer showcase. Shoutout to @OpenAI team for helping with faster generations. Agentic design is here. ✨

Imagine replacing 90% of your employees with a team of geniuses who have no idea how your company operates. Total chaos. Nothing works. That’s what AI feels like today. The missing piece is extracting all the domain knowledge from people’s heads and providing that as structured context to the models.

After months building and forward-deploying @deepintxns , our @ycombinator launch went viral and 4x’d my pipeline. Lesson from 100 AI deployer demos: the hard part is not adoption, it is deployment. We are jumping in with our agents as your teammates, to speed & secure AI, lets go 🚀 deepinteractions.ai

This & ship @deepintxns to prod 🌌 #SF #YC #ragebaiting @edgarpavlovsky

Viral YC launch week as a solo founder = balls everywhere at @deepintxns except @slashyai is in my email slack catching them before i even know i dropped them, ops never felt so easy!!! Go Slashy 💜💚

95% of AI pilots fail. Not because the models are bad. Because teams can't build in sync. Deep Interactions (@deepintxns) is the collaborative AI builder that ships working products in an afternoon. The future of AI isn't more prompts. It's better collaboration. Congrats on the launch, @_sruvis! ycombinator.com/launches/QOt…

How should UIs evolve in the Age of AI? Everyone's declaring GUIs dead. But we're still living through the command-line era of AI — typing instructions into blank text boxes and hoping for the best. open.substack.com/pub/person…

Introducing ml-intern, the agent that just automated the post-training team @huggingface It's an open-source implementation of the real research loop that our ML researchers do every day. You give it a prompt, it researches papers, goes through citations, implements ideas in GPU sandboxes, iterates and builds deeply research-backed models for any use case. All built on the Hugging Face ecosystem. It can pull off crazy things: We made it train the best model for scientific reasoning. It went through citations from the official benchmark paper. Found OpenScience and NemoTron-CrossThink, added 7 difficulty-filtered dataset variants from ARC/SciQ/MMLU, and ran 12 SFT runs on Qwen3-1.7B. This pushed the score 10% → 32% on GPQA in under 10h. Claude Code's best: 22.99%. In healthcare settings it inspected available datasets, concluded they were too low quality, and wrote a script to generate 1100 synthetic data points from scratch for emergencies, hedging, multilingual etc. Then upsampled 50x for training. Beat Codex on HealthBench by 60%. For competitive mathematics, it wrote a full GRPO script, launched training with A100 GPUs on hf.co/spaces, watched rewards claim and then collapse, and ran ablations until it succeeded. All fully backed by papers, autonomously. How it works? ml-intern makes full use of the HF ecosystem: - finds papers on arxiv and hf.co/papers, reads them fully, walks citation graphs, pulls datasets referenced in methodology sections and on hf.co/datasets - browses the Hub, reads recent docs, inspects datasets and reformats them before training so it doesn't waste GPU hours on bad data - launches training jobs on HF Jobs if no local GPUs are available, monitors runs, reads its own eval outputs, diagnoses failures, retrains ml-intern deeply embodies how researchers work and think. It knows how data should look like and what good models feel like. Releasing it today as a CLI and a web app you can use from your phone/desktop. CLI: github.com/huggingface/ml-in… Web mobile: huggingface.co/spaces/smolag… And the best part? We also provisioned 1k$ GPU resources and Anthropic credits for the quickest among you to use.

𝐂𝐡𝐚𝐦𝐩𝐢𝐨𝐧𝐬 𝐨𝐟 𝐭𝐡𝐞 𝐰𝐨𝐫𝐥𝐝, 𝙤𝙣𝙘𝙚 𝙖𝙜𝙖𝙞𝙣 🇮🇳🏆 #T20WorldCup

this started with a striking PC1 falling out of persona space my main insights from the past few months: ⊹ “distance from the Assistant” is the main axis of persona variation across these models e.g. the most relevant thing seems to be “how Assistant-like is this persona” ⊹ this axis already exists in base models and steering with it makes them speak from the POV of helpful archetypes like therapists, coaches, and consultants ⊹ not all personas far from the Assistant are bad! the risk comes from departing the more predictable territory of post-trained behaviour still have a lot of questions about what to anthropomorphize, what to treat as fundamentally alien…

2024 💙💙 2025 CHAMPIONS! #TeamIndia

Introducing Carl, the first AI system to create a research paper that passes peer review. Carl's work was just accepted at an @ICLR_conf workshop on the Tiny Papers track. Carl forms new research hypotheses, tests them & writes up results. Learn more: autoscience.ai/blog/meet-car…

Yes! I just gave a talk last week making the same argument: creating new LLM apps is easier than ever; evaluating their human impact is the new bottleneck of the prototyping loop.

OK, now they have REALLY gone too far.

🚀#AI advances are accelerating, with new models emerging regularly. Benchmark scores only reveal so much. For #HumanCenteredAI, we must ask: How will this model work in my app and for my users? Eureka standardizes LLM evaluation for deeper insights beyond single-score metrics👇

My first student has won the 🏆Best Paper Award with his very first paper! 🎉 Congratulations Han Sanghyeon @Rowhan1029, and a huge thanks to @STAIWorkshop for hosting such an exceptional event in #AIforSustainability at @IJCAIconf