8 projects — from agentic AI systems to reinforcement learning.
A PPO agent learns to play Pac-Man, then an RSSM world model learns to dream the game, and a dream agent learns to play entirely inside those dreams.
Multi-agent AI system using LangGraph and Claude for financial analytics and trade cost analysis.
Pool Apple Silicon Macs into a distributed ML training cluster with auto-discovery, adaptive compression, and thermal-aware scheduling.
Ask questions about your own documents and get answers, entirely offline. No cloud, no data leaving your machine.
Generative Adversarial Network trained on MNIST and deployed as a live demo on HuggingFace Spaces.
Custom GPT-style transformer trained from scratch on BookCorpus using Horovod across 4x H100 GPUs, from raw data to text generation.
Another project is on the way. Stay tuned.
More coming soon. Watch this space.