I don't stop at using the AI stack. I build on it—and fix what breaks underneath.

I'm Gökay, an AI engineer working across LLM infrastructure, autonomous agents, and developer tools. From fine-tuning and RAG at Halkbank to upstream fixes in open-source AI, I work where model capability meets engineering reality.

Build the tool. Trace the failure. Ship the fix.

My projects · Open-source work · X · LinkedIn

Hermes Agent · Nous Research
Agent infrastructure that recovers. Fixed interrupted-update recovery so pending gateway restarts complete on the next run—even when Git already reports up to date. Original contribution incorporated upstream with co-author credit.
View merged work ↗

OpenClaw
Recovery instead of failed turns. Fixed llama.cpp context-overflow detection to enable session compaction and retry. Also repaired plugin-catalog routing for direct navigation and refresh.
Agent recovery ↗ · Catalog routing ↗

Unsloth
Get past installation. Get to the model. Fixed Studio's Linux installation fallback when distro uv policy blocks automatic Python downloads.
View merged work ↗

Cherry Studio
Correctness at the data boundary. Fixed schema validation that broke knowledge-base listing when a database contained zero documents.
View merged work ↗

Open pull requests across inference, training, and agent systems—not yet merged.

• vLLM — KV-cache accounting for hidden-state extraction.
• SGLang — Disaggregated-serving staging, request retraction, and vision preprocessing.
• Hugging Face Accelerate — Avoiding double sharding of already-sharded data loaders.
• OpenHands — Claude Code ACP skill integration and agent-server reliability.
• LMDeploy — Safer HTTP weight updates and validation of untrusted serving inputs.
• LibreChat — MCP instructions, agent generation parameters, and proxy support.
• Jan — Custom OpenAI-compatible providers and llama.cpp runtime configuration.
• Buzz — Bundled search-tool compatibility.

Explore the work →

Agents move fast. Developers should keep control.

Turn-level checkpoints and worktree-scoped restores, written in Rust. Inspect the plan, rewind an agent's changes, and notify it of the restored state. Explicit previews. Scoped restores. Git metadata untouched.

Source · Releases · Architecture

Your agent finished. You should know.

Native completion notifications for macOS and Linux through Claude Code's Stop hook. Built for terminal-heavy workflows, including Ghostty, tmux, and Zellij.

Python · Rust · TypeScript · PyTorch · Docker · Linux

LLM fine-tuning and RAG experience at Halkbank. Studying Artificial Intelligence at the University of Birmingham, graduating in 2027. Building toward agents with better memory, measurable behavior, and recoverable workflows.

Building serious AI infrastructure or developer tools? Let's talk.
X / @gokayai · LinkedIn