Hac Hai Pham

I build modern, maintainable web and AI products end to end. Over 4+ years of production experience with React and TypeScript, I turn business and user requirements into shipped features with strong UX, and increasingly pair that frontend depth with applied LLM engineering.

What I do:

Frontend engineering. React, Next.js, TypeScript, Tailwind. Development and maintenance of the frontend of the Enterprise Data Hub at openpack GmbH inside an Nx monorepo, migration from Webpack to Vite, and delivered core features independently: monitoring dashboards (AmCharts), asset management, mobile views, and an AI-assisted translation feature. Development of the component library in Storybook and safeguarded quality with Playwright E2E tests, Biome, and SonarQube, working together with UX designer and backend developers.

Applied AI / LLM. I design and ship production RAG systems: hybrid retrieval (pgvector + BM25, Reciprocal Rank Fusion, reranking), agents built with Pydantic AI, and evaluation frameworks with LLM-as-judge scoring and Langfuse observability to measure and improve answer quality. Tools I work with: Instructor, LangGraph, Vercel AI SDK, Claude API, context engineering, durable execution.

End-to-end ownership. As lead developer for Maia Earth Organization, I took full engineering ownership of a research platform running in production. It synchronises guided audio sessions with HRV recordings into analysis-ready data, with an interruption-proof mobile session engine and researcher tooling (admin dashboard, multilingual questionnaires) deployed on Render with automated CI backups.

I'm available for freelance frontend and AI/LLM projects with end-to-end ownership. Let's talk if you're building something at the intersection of clean frontend architecture and applied AI.