Hey! I'm an AI Engineer with 2 years of work experience. I specialize in making AI systems reliable in production.
At sensen.ai I worked as a Software Engineer and served as the company's sole Data Scientist. I built a production model-evaluation framework across 26 parking deployments and worked with on-site teams and clients across 4 countries to turn operational failures into measurable engineering problems.
My Master's Research carried the same question into LLM systems. I built a multi-agent Planner-Critic-Fixer pipeline to detect and repair generated-code errors at inference time. My research paper is under conference review.
Lately I'm building an end-to-end system with AI agents for credit card fraud disputes, focusing on LLM evaluation, observability & guardrails.
Company
Aug 2022 to Aug 2024 ยท 2 years 1 month
Jan 2023 to Aug 2024
Sole Data Scientist and forward-deployed technical contact across sensen's client base. Built an ANPR vehicle-direction classifier, a production model-evaluation framework and analytical SQL serving 80 client deployments. Gathered requirements, scoped feasibility and self-deployed 300+ data products on AWS, Azure and GCP servers.
Aug 2022 to Jan 2023
Built 8 ticket-generation APIs on PL/pgSQL functions behind Spring Boot and designed an image-retention policy in Python that reclaimed 400+ GB and deferred server upgrades 4 to 5 months.
AI agent for card-fraud disputes built around evaluation, observability and guardrails. The system connects streamed conversations to transaction lookup, fraud scoring and guarded dispute filing.
Graduate research on whether a Planner-Critic-Fixer monitor improves generated-code correctness across local language models without retraining them.