This is crazy. ml-intern just passed the @huggingface internship test in 15 minutes. The task: replicate a research baseline from a DeepMind…

This is crazy. ml-intern just passed the @huggingface internship test in 15 minutes. The task: replicate a research baseline from a DeepMind paper on test-time compute scaling. Here's what the agent did: - Read the DeepMind paper, dug into Appendix E, picked the right scoring strategy (last-step PRM prediction, not min/product — matters a lot) - Ran 16 solutions per problem through a reward model, grouped by final answer, summed scores, picked the winner - Went from 45% → 65% accuracy. +20pp over greedy. Beat majority vote AND standard Best-of-N - Generated 4 plots, pushed a full results dataset to the Hub, deployed a Docker Space on T4 GPU The funniest part: it cited @lewtun's code snippet as a reference. The intern is already citing your bosses work to look good. It also wrote the README, documented every design decision with paper citations, and added a co-authorship note explaining exactly which parts it did. here's everything it produced: full docs: http://huggingface.co/blog/cmpatino/ml-intern-takehome trained model: http://huggingface.co/cmpatino/math500-bon-exercise dataset: http://huggingface.co/datasets/cmpatino/math500-bon-weighted-results the take home test: http://github.com/huggingface/post-training-takehome 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 …

Source: Clem Delangue (X) | 2026-04-23

Related