Dogukan Uraz T.

RE/RS, Self-Improvement, Program Synthesis and Scaling Laws

---

dtmeai[dot]outlook[dot]comGithubSubstack

I work on the intersection of self-improving AI, program synthesis, neural scaling laws and agent infrastructure for science. I see finding new ways to scale reasoning and improvement as a stepping stone to super-reasoning. Although there are many technical topics that interest me the most are meta-learning, meta-optimization, multi-step reasoning, program synthesis, interpretability, compute-optimal scaling.