← back
The AI Skill I Rely On Daily — Priscila Andre de Oliveira, Sentry
Takeaway
A reusable code-comprehension skill can deliver substantial daily value by making unfamiliar and changing systems easier to understand.
Summary
- At Sentry, a mature codebase with roughly 100 daily merged PRs makes understanding changing conventions and history a continuous task.
- An analysis of 116 assistant sessions classified 67% of usage as comprehension and only 2% as code generation.
- The catch-me-up skill organizes questions into architecture, conventions, feature tracing, syntax, testing, and history.
- Visual structures and targeted explanations accelerate onboarding, regression investigation, and understanding product decisions while preserving human steering.
code-comprehensionagent-skillssentry
Original description
Priscila Andre de Oliveira analyzed 116 of her own Claude sessions from daily work at Sentry. 67% were comprehension. 2% were code generation. Working in a codebase with 15 years of history, around 100 PRs merged per day, and 100,000 organizations depending on it, the unlock is not generation but understanding. She built a personal skill called catch me up with six exploration modes covering architecture, conventions, feature traces, syntax, testing, and history. The loop: understand what the agent found before you let it plan and implement, because a misaligned mental model is where slop comes from. Speaker info: / priscila-andre-de-oliveira-ab34bb24b Timestamps: 0:00 Introduction and speaker background 2:25 Sentry's engineering environment and scale 3:50 AI-driven projects at Sentry 5:48 Maintaining code quality and technical debt 7:35 The role of comprehension in software development 9:38 Analyzing AI usage patterns 10:33 The "Catch Me Up" skill architecture 12:15 Short demo of the "Catch Me Up" skill 13:56 Planning vs. implementation in AI workflows 15:26 Conclusion and key takeaways