Sufiyan Khan

I'm an AI engineer who builds and ships complete AI systems from scratch, end to end — architecture through production deployment.

My flagship project, Groundwork, is an enterprise RAG system with hybrid BM25 + vector retrieval, Cohere cross-encoder reranking, and citation-grounded generation — every answer traces back to its exact source. I evaluated it with Ragas and measured a 1.00 faithfulness score before calling it done. I also built AI Support Pro, a multi-agent customer support platform (LangGraph + AutoGen) that triages, routes, and resolves tickets autonomously, and a real-time voice AI agent using Deepgram STT/TTS and Groq.

Before building independently, I completed an LLM post-training internship at Ethara AI, working inside real RLHF pipelines — evaluating model outputs and feeding structured feedback into production fine-tuning.

I care about full ownership: choosing the stack, making the architecture calls, and being accountable for whether something works in production — not just whether a PR merged. I'm looking for a remote Founding Engineer or AI Engineer role where I can do that at scale, on a team building real AI products rather than thin API wrappers.