What's this "AutoResearch" anyway?

What else can AutoResearch do beyond improving your LLMs?

We will walk through Andrej Karpathy's AutoResearch, a framework that gives an AI agent a real LLM training setup and lets it experiment autonomously. It modifies code, trains for five minutes, checks if the result improved, keeps or discards, and repeats. You wake up to a log of experiments and (hopefully) a better model.

The core idea: you are not touching the Python files like you normally would as a researcher. Instead, you are programming Markdown files that provide context to the AI agent and set up your autonomous research org.

We will start by training an LLM with AutoResearch, then move quickly into other use cases. Thousands of developers worldwide have already taken this library well beyond its original scope. We will explore what they found.

autoresearch is an open-source library by Andrej Karpathy, former head of AI at Tesla and founding member of OpenAI, built to automate training/researching of GPT-style models as simple as possible.




Note:

- Lunch will be offered.
- Students should never miss class to attend optional programming. The Deans’ Office will not support requests for absences due to participation in any non-class activity or Booth-sponsored event as excused.

Artificial Intelligence [AI] Club (E/W) | Website | View More Events

Contact the organizers