Hi there, I am a Data and ML engineer with a Mathematics and Data Science foundation from UC Berkeley. I've shipped production RAG pipelines and agentic applications, and I care as much about proving they work (evaluation, hypothesis testing) as building them.
Currently: M.S. Business Analytics @ Santa Clara University · San Jose, CA
B.A. Cal Alumni · Berkeley, CA
A multi-agent retrieval-augmented generation pipeline. Rather than a single retriever-plus-LLM loop, JJRAG breaks complex questions into sub-queries, routes each to a specialized agent, and reconciles the results into one grounded answer. The design came out of a problem I kept hitting with financial and technical documents: sources are inconsistently structured, and naive chunking loses the semantics that make retrieval useful. JJRAG uses schema-first extraction so the structure survives ingestion.
A conversational trip planner. Most itinerary tools ask you to fill out a form; this one has a conversation. It elicits constraints incrementally — where, when, how much, what you actually care about — and holds that state across turns while calling out to live location and routing data.
An RNN language model that predicts subsequent phrases from dynamic text input, combining LSTM and GRU architectures to solve vanishing-gradient limitations and extend long-term contextual memory.
I'm open to Data Science, ML engineering, Analytics, and Research opportunities. The fastest way to reach me is email — I read everything.
jinfengj12ang AT gmail DOT com