Jose Esparza

I like owning a system from the first prototype to the dashboards that watch it in production. As a Senior ML Engineer, most of what I ship now is LLM systems: agentic assistants with LangGraph and MCP, RAG over vector databases, and fine-tuned Llama models for domain-specific generation, plus the BI and observability around them. I've worked across fintech, edtech, retail and large enterprise.

Earlier in my career I led a five-engineer data team building AWS pipelines that ingested product-availability data from Cencosud, Falabella and Walmart, and trained default-risk models for fintech lenders. Before that I engineered raw-material microservices for INDITEX, refactored the Spanish Ministry of Justice's certificate system to handle 500k+ records, and shipped a full-stack risk platform for Barrick Gold. My resume has the specifics.

I'm deep in the Julia open-source community. Google Summer of Code 2023 funded my work on empowering Julia-based Data Science with MLflow. I contribute to the Machine Learning in Julia (MLJ) framework and maintain DearDiary.jl under the JuliaAI organization.

I hold a B.S. in Software Engineering from Universidad Privada del Norte, where I graduated with a 4.0 GPA. At Skolkovo's SMILES 2020 summer school I placed in the top 10% of 2,000+ applicants and presented my work on Convolutional Neural Networks for respiratory-condition detection. I'm pursuing an MSc in Computer Science at the University of London.

As a REPU Computer Science alum, I analyzed 80,000 COVID-19 texts with Gisella Bejarano in Arti Ramesh's lab at Binghamton University, and we published a paper indexed in the WHO COVID-19 Research Database.

Beyond the code, I do vocals in a band called El mejor verano de mi vida (The best summer of my life). We got counted in the fifth wave of emo, and I'm proud of it. I also play Magic: The Gathering and spend time with my two cats.