hey, i'm dt. i'm a researcher who loves working on recursive self-improvement, scaling RL compute & building RL training pipelines for autonomous math & code-reasoning agents. i'm currently focused on learning, exploring & experimenting with what's possible when AI thinks & acquires more compute adaptively on longer problems. i love pushing RL compute scaling for LRMs that can autonomously turbocharge code generation.

my main quest is practicing with proto-super-innovator agent systems that accelerate science & code generation on problems that actually matter—or even on research itself. i'm out here building prototypes & sharing my work with the open-source community.

RESEARCH_LOGS

ARCHIVE