Selected research, systems, and prototypes I've built or led.
A 4B medical reasoning model from Typhoon and SiData+ that generates ranked diagnoses, capturing clinical uncertainty and outperforming larger models on major medical QA benchmarks.
A benchmark and framework for evaluating how well LLMs follow structured output formats in narrative PCG tasks, with error taxonomy and scaling analysis.
Contributed to OpenRLHF, an easy-to-use, scalable, and high-performance RLHF framework.
Interactive timeline visualizer for Thailand earthquake events with filtering and temporal context.
Lightweight evaluation platform for LLM experiments.
A GUI program for labeling datasets.