Full-Stack AI Engineer. I build the model and the product around it.
Most ML engineers stop at the model. I ship the whole thing: the AI, the backend, the frontend, and the payments layer.

What I build:
Generative AI (image and video): Orvia Labs, a multi-provider AI image generation studio integrating DALL·E (OpenAI), Stable Diffusion 3.5 Large Turbo (Hugging Face), Imagine Art (Vyro), and Gemini Image behind a single FastAPI backend and React studio interface, with per-model controls, prompt improvement, and flexible dev/production deployment. I also work with text-to-video models including Imagine Art (Vyro), Kling AI, and Runway, and have built text-to-image and text-to-video generation pipelines in production.

Computer Vision and Medical AI: Dental AI, an end-to-end SaaS platform for dental X-ray analysis (EfficientNet-B4, YOLOv11, U-Net) with disease detection, implant feasibility prediction, automated clinical reports, a RAG chatbot, and Stripe subscription billing.
Full-Stack SaaS: Orvia Studios, a deployed marketing agency platform with an authenticated client portal and admin dashboard, role-based access control, campaign analytics, and invoicing. Live at orvia-studios.vercel.app.

Production experience: Around two years building AI systems for international clients (currently a UK music label, previously a Canadian startup), including multi-modal content moderation pipelines (Whisper, EasyOCR, classification models) deployed as scalable FastAPI microservices.

Stack: Python, FastAPI, PyTorch, TensorFlow, React, Next.js, MongoDB, PostgreSQL, Docker, AWS. LLM APIs (OpenAI, Claude), RAG, embeddings, vector databases. Claude Code daily.

I share plans up front, communicate proactively, and ship. Happy to start with a small paid task so you can judge my work rather than my pitch.