Visiting Engineer - ContextHub

Here’s what you’ll do

• Own Context Hub end to end. You'll be the single-threaded leader for the product — responsible for its technical direction, growth strategy, and community health. You won't be handed a roadmap; you'll create it.
• Talk to developers and understand what they need. The best product decisions come from deep empathy with users. You'll engage directly with the developer community — through GitHub issues, user interviews, and usage data — to identify what to build next and why.
• Invent new ways for AI agents to discover and use knowledge. Context Hub started as a CLI for fetching docs, but there's enormous room to push further: smarter retrieval, agent-to-agent knowledge sharing, self-improving documentation, private team contexts, and ideas we haven't thought of yet. You'll explore the frontier.
• Build and maintain the Context Hub CLI and core infrastructure, ensuring reliability and performance as usage scales across thousands of developers.
• Design and implement systems for intelligent documentation retrieval, versioning, and curation so that agents always get the most accurate and relevant context.
• Develop features that allow agents to annotate, enrich, and share knowledge across sessions—building toward a self-improving documentation ecosystem (the “Stack Overflow for AI coding agents” vision).
• Collaborate with the open-source community to review contributions, triage issues, and maintain a high-quality contributor experience.
• Experiment with LLM-based approaches to automatically validate, summarize, and improve documentation quality.
• Instrument, monitor, and optimize the tool’s performance and developer experience based on real-world usage patterns.
• Work closely with Andrew Ng and the engineering team to define the product roadmap and technical direction for Context Hub.
• Collaborate with internal teams to create content, processes, and AI-powered solutions for scalable support of the developer community around Context Hub.

Here are the skills you should have

• Product sense for developer tools. You understand what makes developers adopt (and love) a tool. You've shipped something that other developers use — whether it's an open-source project, a CLI, an SDK, or an internal platform — and you made decisions about what to build, not just how to build it.
• Community instincts. You know how open-source communities work — what motivates contributors, how to write docs that people actually read, and how to turn users into advocates. You're comfortable being a public face for a project.
• Strong proficiency in TypeScript/JavaScript and Node.js, with experience building and shipping CLI tools, developer SDKs, or developer-facing infrastructure.
• Very strong understanding of LLMs and AI coding agents—you’ve built applications that integrate with or orchestrate language models, and you stay current with the rapidly evolving AI tooling landscape.
• Experience with npm package publishing, versioning, and open-source project maintenance.
• Strong systems thinking: you can design APIs, data models, and architectures that are clean, extensible, and well-documented.
• AI fluency: Strong conceptual understanding of AI trends and technologies. Familiarity and experience with AI-powered coding tools such as Claude Code, Codex or similar—having built or prototyped software with AI assistance.
• Entrepreneurial mindset: Self-starter comfortable working in a fast-paced, small-team environment where you own projects end to end and thrive in ambiguity.
• 3+ years of professional software engineering experience.

Bonus if you have any of the following

• Experience maintaining a popular open-source project (1,000+ stars).
• Background in information retrieval, search, or knowledge graph systems.
• Experience with MCP (Model Context Protocol), tool-use frameworks, or agent orchestration platforms.
• Familiarity with documentation-as-code workflows and static site generators.
• Graduate degree (Master’s or PhD) in Computer Science, AI, or a related field.