ML Infrastructure Engineer. Building AI generation systems at scale.
Building and scaling AI image generation infrastructure serving ~10K daily users. Working with models like SD1.5, SDXL, Flux, Wan, Hunyuan architectures.
Core focus:
- Inference optimization (achieved 20x speedups with techniques like Sage Attention)
- Model integration & deployment (ComfyUI, Stable Diffusion ecosystem)
- Production ML pipelines with CUDA/PyTorch
- Multi-agent AI systems & orchestration frameworks
ML/AI: PyTorch • CUDA • ComfyUI • Stable Diffusion • Flux • Wan • Hunyuan
Infra: Docker • Redis • PostgreSQL • RunPod
Languages: Python • Rust • TypeScript • Go • Bash
Easy Dictate — Desktop voice-to-text app with AI transcription. Push-to-talk dictation with OpenAI Whisper, Groq, and ElevenLabs. Built with Tauri v2 + Rust.
- AI agent orchestration
- Zero-error task execution frameworks
- Autonomous AI systems architecture
- Advanced inference optimization techniques
Open to collaborations on ML infrastructure and AI optimization projects.