Papers & projects
Research papers, technical reports, systems work, and open-source contributions across agentic LLM post-training, inference, CUDA, reinforcement learning, and video generation.
Occamy-1.0
Open Pareto-frontier 35B Intelligence for Co-work
CineFlow
Co-first author — dependency-driven parallel execution for scalable, consistent video generation
RareDx
Co-first author — controlled knowledge integration and graph-grounded policy optimization for rare-disease diagnosis
Open-Source Systems Engineering
RL infrastructure, long-context training, inference optimization, and reliable training pipelines
Dynamic Prefill Optimization
Adaptive online packing for lower-TTFT LLM serving
FlashAttention-style CUDA Optimization
Memory-efficient kernels and fusion for GPT-2 inference
Reinforcement Learning for Legal Reasoning on Multi-Choice QA
Hybrid Zero-RL, distilled-CoT SFT, and GRPO for structured legal reasoning