The AI Skill I Rely On Daily — Priscila Andre de Oliveira, Sentry

2026-05-27 · Source: AI Engineer · Field: Technology & Digital — Software Development & Engineering, Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems · Depth: Intermediate, long

Summary

Priscila Andre de Oliveira, a Senior Software Engineer at Sentry, highlights that her primary daily AI skill is code comprehension, not generation. Despite Sentry's extensive AI initiatives, including tools like Abacus for AI usage tracking, Warden for code reviews, and Junior for Slack-based bug resolution, Oliveira's personal AI usage analysis revealed 67% of her prompts focused on understanding code, with only 2% for generation across 116 Claude sessions. She developed a local "Catch Me Up" AI skill, a detailed prompt structured into six exploration modes (Architecture, Convention, Feature Trace, Syntax, Testing, History), to rapidly gain context on Sentry's complex, 15-year-old codebase, which sees 100 daily PR merges. This skill helps her quickly grasp new repositories and review PRs, emphasizing that aligning one's mental model with the code is crucial for effective AI steering and preventing "slop code."

Key takeaway

For software engineers navigating complex, evolving codebases, prioritize AI for deep comprehension rather than just code generation. Track your own AI usage to confirm this pattern, as understanding the existing system is crucial for effective AI steering. Develop custom, structured prompts or "skills" to rapidly gain context on architecture, conventions, or history. This approach ensures you align your mental model with the codebase, preventing "slop code" and enabling you to ship high-quality, "keynote" code, as comprehension is AI's most significant contribution.

Key insights

AI's primary utility in large, complex codebases is comprehension, significantly outweighing code generation.

Principles

• Code comprehension is paramount for effective AI steering.
• Custom, structured prompts enhance AI's analytical capabilities.
• AI can act as a tireless teammate for context acquisition.

Method

Develop a "skill" by creating a detailed, goal-oriented prompt (e.g., an MD file) that structures comprehension questions into specific exploration modes like architecture or history.

In practice

• Analyze personal AI prompt history to identify usage patterns.
• Implement custom AI skills for rapid codebase context acquisition.
• Employ AI to accelerate understanding during PR review processes.

Topics

• AI Code Comprehension
• Prompt Engineering
• AI Agents
• Sentry Platform
• Software Observability
• Technical Debt Management

Best for: Software Engineer, AI Engineer, Prompt Engineer

Related on AIssential

• My cheatsheet for a clean context — Effective coding agent use requires mindful context management, precise prompting, and systematic debugging.
• 5 Brutally Powerful AI Prompts I Wish I Knew a Year Ago — Effective AI interaction stems from structured prompt design, not just advanced tools or models.
• AI Coding Tools Raise the Ceiling for Developers, Not Replace Them — Technical depth is a force multiplier for developers leveraging AI, ensuring correctness, sustainability, and effective intervention.
• This Skill Makes AI Coding Work: Navigating Context Engineering in 2026 — Context engineering is the pivotal skill for effective AI-assisted coding in 2026.
• Faster Code, Slower Understanding: A Survival Note from a Year of Agent-First Engineering — AI shifts the engineering bottleneck from code production to comprehension and maintenance, demanding new optimization strategies.
• Beyond Prompting: How Engineers Should Architect AI Code Generation — AI integration in software development shifts from prompting to designing controllable, continuous production processes and workflows.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by AI Engineer.