← back
GitHub Next Explorations: Rahul Pandita
Takeaway
GitHub Next bets on staged exploration: code-completion is evolving to multi-location 'Next Edit Suggestions' and to full inner-loop task completion via Copilot Workspace.
Summary
- GitHub Next is ~20-person research team reporting to CEO; incubated Copilot, Copilot CLI, Copilot Voice, GitHub Blocks, SpecLang etc.
- Compares AI transition to 1880s-1920s electric motors: 40 years of factory-floor exploration before mainstream
- Copilot Next Edit Suggestions extends ghost-text to multi-location edits (e.g. add param → method def → call sites → docstring); fine-tuned models, exploring UX for out-of-viewport edits
- Copilot Workspace targets the inner loop (task→spec→plan→implementation), emerged from merging SpecLang and runtime-aware prompting explorations
copilotgithubcode-generation
Original description
In this session, we'll talk about current explorations by GitHub Next including Copilot Workspace a copilot native dev environment for getting started with everyday tasks. Recorded live in San Francisco at the AI Engineer World's Fair. See the full schedule of talks at https://www.ai.engineer/worldsfair/2024/schedule & join us at the AI Engineer World's Fair in 2025! Get your tickets today at https://ai.engineer/2025 About Rahul I am Rahul Pandita, a Staff Researcher at Github. My primary research interests are in data science and automated software engineering. I specifically work in the area of program comprehension targeted towards improving developer/tester/end-user productivity. Previously, I worked as a Senior Researcher at Phase Change Software. Before that, I was as a postdoctoral researcher at Department of Computer Science NCSU working with Dr. Laurie Williams and Dr. Munindar Singh. I received my PhD. from NCSU in Computer Science. And I believe that “There is so much more left to be done...”. I strongly believe that there is always room for improvement.