AI in the SDLC: a team workshop

For an engineering team adopting AI across the build process (SDLC, the software development life cycle: idea to shipped, tested code) without losing control, code quality, or its own understanding of the system.

The failure mode is quiet: code gets merged that nobody fully understands, review gets skipped, architecture drifts. Speed today; an unmaintainable black box tomorrow.

What the workshop is

Hands-on, on your own codebase. Where AI fits the lifecycle, and where it doesn’t. At its core is a five-phase delivery method with two mandatory human review gates, design review and merge review, that never get automated away. The gates are where control lives.

Formats

• Half-day executive briefing. Frame, risks, plan; for leadership buy-in.
• Two-day hands-on workshop. The flagship, run on your team’s real code; about 70% practice.
• Multi-week cohort. The same spine, with practice between sessions.

What your team keeps

A participant workbook, prompt-template cards, an AI-usage policy, a risk register, a 90-day adoption plan. If you take the implementation option, a working GitHub setup as well: board, issue templates, validation checks, and agent automation within the limits you set.