​AI-Memory Hackathon: Building your own Agent LLM Wiki 🚀

​"Instead of just retrieving from raw documents at query time, the LLM incrementally builds and maintains a persistent wiki" - Karpathy

​Recently, Karpathy popularised the idea of LLM Wiki. We decided to give it a go and try to extend on that idea and build one together with you.

​​​​🏆 1,200€ Cash Pool

​Main Prizes:

​🥇 Prize: 600€ cash
🥈 Prize: 400€ cash
🥉 Prize: 200€ cash

​​🧰 Build LLM Knowledge Wiki

​In this hands-on hackathon, you will use Cognee’s open-source memory engine to build your own LLM Knowledge Wiki

​Your wiki should do three base operations:

1. ​Ingest

2. ​Query + Self-improve

3. ​Lint

​Check how Karpathy explains it here: https://gist.github.com/karpathy/442a6bf555914893e9891c11519de94f

​​💥 Show us what you got

​You will have 3 minutes to stand out:

• ​Present your idea and explain how you leverage agent self-improvement

• ​Run a live demo that showcases your agent in action

​​​​⏱️ Schedule

• ​17:00 PM — Doors open + Networking

• ​17:30 PM — Opening remarks + Redis & Cognee demos

• ​18:00 PM — Hacking begins

• ​8:30 PM — Project Submission Deadline - finalists selection

• ​9:00 PM — Finalist presentation & judging

• ​9:30 PM — Awards ceremony

• ​10:00 PM — Event wrap-up & doors close

​Limited spots available. Can't make it? Please cancel your RSVP so we can offer your place to someone else.

​About Cognee:

​Cognee is an open-source AI memory engine that transforms unstructured data into persistent, queryable memory. It enables AI agents to store, retrieve, and reason over information across time, making it possible to build more reliable, adaptive, and stateful AI systems.