Building at the intersection of real-time systems, generative AI, and interactive environments.
I specialize in low-level system programming, cross-platform audio capture, shader development for VRChat, and AI-driven pipelines. My engineering philosophy: performance, clarity, and reliability.
Languages
Frameworks & Tools
Cloud & Infra
AI / Audio / XR
| Project | Description |
|---|---|
| 0xBitNet | Run BitNet b1.58 ternary LLMs with WebGPU — in browsers and native apps. |
| PyGPUkit | Minimal GPU runtime for Python — high-performance CUDA kernels, memory management, and LLM inference without heavy dependencies. |
| ProcessAudioTap | Cross-platform per-process audio capture (Win/macOS/Linux). Low-latency taps, real-time filters, plugin architecture. |
| AudioLink Shader Suite | Custom VRChat shaders reacting to audio spectra. Used in club events, DJ worlds, and visual installations. |
| LLM Automation Pipeline | Whisper → LLM → Slack/Discord. Diarization, summarization, topic detection, automated posting. |
| VRChat Automation Tools | Avatar detection, join-log analysis, OSC automation, and AudioLink-reactive effects. |