Presentation
Why work with me?
- Production-First AI: I don’t just build models; I deploy them. My expertise includes fine-tuning LLMs and Computer Vision models (YOLO, PaddleOCR) to achieve 95%+ accuracy in real-world scenarios.
- Cost-Efficiency Specialist: I specialize in model distillation and quantization, allowing complex AI to run on consumer-grade hardware (4GB-8GB VRAM). This reduces your dependence on expensive cloud GPUs and lowers your monthly infrastructure spend.
- Full-Stack Reliability: With a background in Rust (Tauri) and TypeScript (Next.js), I build the robust architecture needed to support AI features. I lead the maintenance of an open-source tool with 65,000+ downloads, ensuring hardware stability across Windows and macOS.
