Incorporating AI into your SDLC
summary
Practical roadmap for using AI across the SDLC: index code and docs into a vector DB + RAG to answer planning questions like “where do we use dependency Y” (often returns unexpected counts, e.g. 10 files), generate prototypes and estimates from meeting comments, auto-create unit tests and pull requests, and run GitHub Copilot agents to suggest changes and iterate until tests pass. Emphasizes custom review/test instructions, treating AI as an intern that must be reviewed, and never skipping human validation.
AI-written summary · about
chapters (20)
- 0:00Introduction and speaker intro
- 1:09Speaker background and experience
- 3:18Three modes of using AI assistants
- 6:09AI in SDLC planning phase
- 8:13Prototyping with AI in planning
- 9:17AI tools for development phase
- 11:03AI-assisted coding tasks
- 13:27AI as an intern-like assistant
- 14:32Agent functionalities in AI tools
- 15:16Resolving issues with AI insights
- 16:41AI for summarizing and tracking
- 17:23AI code review capabilities
- 18:26AI in testing and debugging
- 20:34AI for refactoring and updates
- 21:38AI-driven documentation generation
- 23:03AI for codebase onboarding and quizzing
- 23:45Q&A: Teaching AI to students
- 25:47Q&A: Large codebases and context limits
- 27:51Q&A: AI tool alternatives to Copilot
- 29:57Closing remarks and thanks
related talks
30mBuilding Tools with Tools: Modern AI-Driven Development
26mAI-augmented SDLC: Rethinking how we deliver software
15mAI's Impact on Development, Post-Production & Support Journeys
37mAI-assisted development: Common pitfalls and smarter practices
18mUsing AI Across the Dev Lifecycle
54mGen AI & the SDLC: Changing the way we bring digital products to life
26mAccelerating DevOps with AI
34mAI and Code Quality: Building a Synergy with Human Intelligence