Tackling tech debt
summary
Claims that AI agents let teams clear backlog items faster with less manual work and demonstrats it by assigning a GitHub issue to GitHub Copilot’s coding agent, which creates a pull request in about ten minutes, adds unit tests, and includes the user-event testing library. Presents a live demo app (commit graph gan) where the agent generates tests, shows PR diffs and workflows, and recommends the W.R.A.P issue format, atomic tasks, and time blocking to reliably reduce tech debt.
AI-written summary · about
chapters (15)
- 0:00Introduction: Meet Brittany Ellich and GitHub Copilot
- 0:43Agenda: Tackling Tech Debt with AI
- 0:56The Tech Debt Graveyard: Why It Happens
- 2:20Consequences of Ignoring Tech Debt
- 2:47Introducing the Tech Debt Elimination Accelerator
- 3:00How GitHub Copilot Coding Agent Works
- 4:02Demo: Automating Unit Test Creation with Copilot
- 7:03Reviewing Copilot’s Pull Request Output
- 8:12Building a Sustainable Tech Debt Strategy
- 8:54W.R.A.P Framework: Write Effective Issues
- 10:28Refining Instructions for Better AI Results
- 11:41Breaking Down Large Tasks into Atomic Chunks
- 12:50Pairing with the Coding Agent: Human + AI Collaboration
- 14:42Time Blocking and Workflow Tips
- 15:12Final Thoughts: Turning Tech Debt Mountains into Managed Terrain
related talks
52mEnsuring Quality Testing in an AI-driven World
26mWhen Agile and AI Collide
33mEnsuring quality in an AI-driven world
09mImproving code reviews with AI agents
24mGitHub Copilot Agents + Sonar | The Blueprint for Agentic Development
41mImproving DevOps workflows with GenAI
46mRecalibrate - How AI shakes up software team
50mHow GitHub operationalizes AI for team collaboration and productivity