Harrison Chase Portfolio
Creator of LangChain; his GitHub profile (10,000+ followers, 68 repositories) shows the early, individually-shipped projects — including a Q&A-over-Notion tool and a "chat with your data" app — that preceded and fed into LangChain becoming the most widely adopted LLM application framework.
What makes it work
A breakdown of the choices that make this portfolio stand out.
Small, individually shipped projects visible before the big one
Repos like notion-qa and chat-your-data are complete, scoped applications in their own right — the public history shows the pattern-finding that led to a framework, not just the framework appearing fully formed.
Star counts as an honest, third-party signal
Visible star counts on individual repos (2.2k, 972, and more) give a visitor a real, unmanipulated read on which specific ideas resonated, rather than a self-reported claim of impact.
GitHub itself as the portfolio, with no separate personal site needed
For an open-source-first engineer, a well-maintained GitHub profile is a complete portfolio on its own — proof that the platform where the work lives can double as where it's showcased.
What AI Engineers can take from this
Specific, actionable tips to apply to your own portfolio — no generic advice.
Ship small, complete, narrowly scoped projects publicly before attempting something ambitious — the pattern across those projects is often what the bigger idea comes from.
Let star counts and forks speak for themselves as an honest, third-party signal rather than only self-describing your project's impact.
If your work is open source, treat your GitHub profile itself as your primary portfolio rather than building a separate site to describe the same repos.
More AI Engineer portfolios
Ready to build your portfolio?
Follow the AI Engineer roadmap — skills, projects, and timeline to get hired.