Applied AI Engineer · Full-Stack
Building production AI systems from the ground up — not wrappers, not tutorials. RAG pipelines, LLM integrations, and transformers trained from scratch. Every line ships to production.
Full transformer architecture implemented in raw PyTorch — zero pretrained weights, zero copied code. Multi-head self-attention, causal masking, positional embeddings, GELU FFN — all written by hand. Trained 98,000 steps with mixed precision and cosine LR scheduling.
Production full-stack marketplace with hybrid AI recommendation engine — vector similarity + LLM reasoning, live with real users.
Modular FastAPI backend with RAG pipelines for context-aware AI responses over enterprise knowledge bases.
WordPress plugin submitted to WordPress.org — generates product content via Claude, GPT-4, Gemini, and LLaMA with BYOK model selection.
Open to AI engineering and SDE roles, remote internships, and interesting projects worldwide. If you're building something — let's talk.
Send Message