Empowering OSS maintainers in the age of AI
summary
Apache Airflow sees a jump from ~100 to ~500 open PRs and security reports rise from ~10 to 30+ per month, so a skill‑based agent stack triages, labels and closes issues fast enough to drop untriaged counts dramatically; agents run as parallel “interns” (author runs eight concurrently), skills are plain‑English instructions with Python deterministic hooks that produce JSON CVE records for Vulnogram, auto‑PR fixes, and form the basis of the proposed Apache Magpie project to automate PR triage, release chores and first‑reader onboarding.
AI-written summary · about
chapters (12)
- 0:00Introduction and background
- 3:20AI agents overwhelming maintainers
- 5:52Using agents for repetitive tasks
- 9:20Human judgment in open source
- 13:32Revolution in code review process
- 15:43Concurrency with AI agents
- 19:34Human interactions with agents
- 23:47Introducing Apache Magpie
- 27:42Skills as English instructions
- 31:53Intuition and agent workflows
- 34:40Community over code remains key
- 36:49Q&A session
related talks
01h 01m"Looks Good to Me": A Practical Guide to Handling AI-Generated Code
43mWriting for Machines: How To Capture Your Project's "Vibe" and Survive AI Slop
14mAI's impact on open-source contributions and mentoring
21mHow To Make AI-Assisted OSS Contributions Review-Ready
52mScaling the Security Researcher to Eliminate OSS Vulnerabilities Once and for...
32mFrom Toil To Triumph: Harnessing Agentic AI To Streamline Infrastructure as Code
21mThe OSS maintainer to staff engineer pipeline
35mManaging the AI Surge: Governing & Scaling in the AI Era