I'm an AI/ML engineer focused on LLM training and post-training, reinforcement learning, agent systems, evaluation, inference optimization, and AI infrastructure.
I've independently built and open-sourced several systems, including Delta, a coding-agent harness; GraphWeave, a hybrid graph-vector retrieval system; and RLMness, a recursive LLM execution runtime.
My work spans research-driven experimentation, model optimization, backend engineering, and building reliable AI systems from scratch. I'm particularly interested in research engineering, applied AI, ML infrastructure, and roles involving challenging technical problems with meaningful ownership.
