Dogukan Tuna profile photo

Dogukan Tuna

Hi, I work on sample-efficiency, continually learning AI, and RSI mechanisms aimed at superintelligent / superhuman capabilities. A lot of what I’m exploring involves C/CUDA RL environments and multi-agent architectures. I share live worklogs here, and you can find my open-source projects on GitHub.

Reach out:

Find me:

Work

What I tinker with day to day

  • Speedrunning

    Mythein LabsMachines that get better at getting better.

    Sample-Efficiency · RSI · C/CUDA RL

Live Worklogs

Raw notes from my learnings, builds

Open logbook →

Nothing here yet — the first note lands soon.

Writing

Research notes, build logs, and the occasional deep dive

All 9 posts →
  1. Multi-agent orchestration with Codex, borrowing Fable 5.1 and Fugu Ultra v1.1Codex
  2. Verified Replay Distillation (VRD) recipe for continual learning in verifiable domainscontinual learning
  3. autoresearch-mamba: Karpathy-Style Autoresearch for Mamba-2, Mamba-3, and Hybrid Mamba–Transformer MoEMamba
  4. A self-improving skill catalog for AI agentsGPU programming
  5. Mem-RLM: Memory-Augmented Inference for Recursive Language ModelsLLM inference
  6. Claude Code-Time Skill Acquisition with Agent TeamsAI agents