AgentCon / MCPCon: AI and Open Source Mentorship

Overview

Earlier this year I ran a cohort with over 100 mentees focused entirely on open-source contributions and emulating a real-world developer workflow. There were 14 teams, participants sent over 16,000 messages in the Discord, and 651 contributions were reviewed. Then I ran an exit survey.

The results were telling: 71% of participants used AI to augment their work, and 58% wished I had provided better guidelines around how to do so. That last metric was the lowest-rated portion of the entire experience.

This talk shares the boots-on-the-ground findings: what the data show, how AI helped participants stay engaged, and how it made it harder to identify skill gaps that led to contributor churn.

Key Takeaway

AI is drastically reshaping the open source ecosystem. It is the core driver of the gap between contributors who stay and contributors who churn. As MCP-enabled tooling continues to redefine what "AI-augmented workflow" means, mentorship becomes more important than ever — and the lessons from this cohort belong to every maintainer.

We need to adapt to the ever-changing agentic AI domain. The talk ends with concrete guidance on how to do that.