Staff Software Engineer, GitHub Intelligence (Copilot Agents)

About GitHub GitHub is the world’s leading platform for agentic software development — powered by Copilot to build, scale, and deliver secure software. Over 180 million developers, including more than 90% of the Fortune 100 companies, use GitHub to collaborate, and more than 77,000 organisations have adopted GitHub Copilot. Locations In this role you can work from Remote, United States Overview GitHub is shaping the future of software development in the age of AI agents. The difference between a general-purpose agent and an expert is often whether it can find, understand, and apply the right context at the right time.     One of our goals is to make agents more effective and efficient by default, without requiring every developer or team to become an expert in configuring AI systems.  We're looking for a Staff Software Engineer to help  ideate, develop, experiment, and build Git H ub ’ s context layer that is part of our agent platform – this includes agent memory, search, GitHub graph, bringing the right context to agents efficiently and at the right time.      This work will help  agents  use the context available across code, issues, pull requests, workflows, connected developer tools , and much more . It will also contribute to agentic experiences that can learn from prior work, make higher-quality decisions, and carry useful context across tasks and product surfaces.     You'll  be part of our Copilot Agents organization, working with a strong group of engineers and partnering across product, engineering, design, research, and data science. Your work can positively  impact  millions of developers and the AI agents helping them build everything from open-source projects to global enterprises.     We are looking for creative problem solvers and diverse thinkers, people who care about culture as well as customers and features. We believe that how we do things is as important as what we do. Big vision, a common purpose, passion for quality , moving  fast  and learning quick.  Great products reflect the teams that build them.     The ideal candidate has deep experience designing and  building agent harnesses, evals, developer tools , context retrieval,  and has worked on one or more of information retrieval, knowledge or context systems, developer platforms, or production generative-AI applications. You can move from ambiguous product problems to clear technical direction, build end-to-end systems, and help multiple teams make coherent architectural choices. You learn quickly, communicate clearly, and combine experimentation with operational rigor.

Provide technical leadership for systems that help AI agents acquire, interpret, retain, and apply relevant software-development context. Design, prototype, implement, test, ship, and operate high-quality technical solutions that scale across GitHub services and Copilot experiences. Develop platform capabilities and APIs for context retrieval, sessions, memory, agent customization, evaluation, and feedback-driven improvement. Turn ambiguous customer and product opportunities into clear technical strategies, architecture proposals, experiments, and executable roadmaps. Work at the frontier of how software gets built with agents and evolve workflows that make the whole team faster.  Partner with product managers, designers, researchers, data scientists, and engineers across GitHub to connect capabilities into coherent end-to-end experiences. Establish evaluation methods, telemetry, and learning loops that measure agent quality, efficiency, reliability, and customer outcomes. Make thoughtful tradeoffs across capability, latency, cost, privacy, security, reliability, and developer experience. Own and advocate for the health and quality of the systems the team builds, including participating in first-responder on-call rotations and live incidents where applicable. Apply debugging tools, operational data, and structured