ls ./agentic-systems
Building agentic systems that act ethically and with purpose.
Multi-agent orchestration of physical robot systems. This project explores building a fully autonomous robot orchestrated using multi-agent LLMs, enabling humans to interact with it more naturally, see how it reasons, and watch its understanding evolve in real time.
Built to experiment with AI that can meaningfully interact with the physical world.
A local-first web app that turns a thinking session into a living graph. Notes are staged as cards in collaboration with an agentic thought partner, then embedded and clustered on a graph canvas to reveal semantic similarity, blind spots, and metacognitive insights.
Created to explore how AI can help capture thoughts temporally and act as a genuine thinking partner rather than an autocomplete or QA assistant.
Autonomous quadcopter that detects and tracks targets based on natural language commands over an LTE connection. Real-time video is streamed from a Raspberry Pi over cellular.
Built to test natural language control of physical systems using language models and computer vision despite constrained hardware.