On AI Lounge since Nov 2026
Keylhan Paumard
Member #4,534DeveloperL2AI/ML Student, Independent
ML EnjoyerStudent @ EFREI Paris
How karma works. Everyone starts with 5. You get +1 each time another member upvotes one of your posts or comments (your own upvotes don't count), and lose it if they take the upvote back. Post what you're building, answer questions and help other members to earn more.
Achievements
Plugged In ×3Connected sources: GitHub, Hugging Face, LinkedIn, X, website.How: Connected 3 sources (GitHub, LinkedIn, X or a website) in Settings
Plugged In×3Featured
ShipperProjects on your profile.How: Added 1 project to their profile
Shipper
No posts yet
Keylhan Paumard hasn't posted yet.
Pinned projects
At-home reproduction of Exa-style web-scale neural search: binary-quantized IVF vector DB, LUT scoring, cross-encoder rerank, low-latency MCP — a learning build.
Minecraft video world model (AR transformer + SD-VAE) - pre-cleanup snapshot, moving to local RTX PRO 6000
Security guards, disciplined multi-agent orchestration, and persistent memory for Claude Code
About
Keylhan Paumard is an AI and ML student at EFREI Paris who works independently from Paris, France. Known on GitHub for llamaup, a set of pre-built Linux CUDA binaries for llama.cpp organized by GPU architecture, and for exa-at-home, a hands-on learning project that recreates Exa-style neural search at home. Keylhan is currently developing minecraft-world-jepa, a Minecraft video world model, and zenno, which adds security guards, multi-agent orchestration and persistent memory to Claude Code.
