I was recently a guest on Off Script, a podcast hosted by Josh Nesbitt and James Hall. Episode 60, released on 15 April 2026. Thirty-nine minutes.
We talked about using AI as a practical tool for the less glamorous parts of engineering: refactoring old codebases, keeping on top of tech debt, and making documentation actually useful. A lot of the ground the conversation covered overlaps with things I've been writing about here — particularly the idea that AI's most interesting contribution isn't the new features it helps you build, but the legacy weight it helps you shift.
If that's something you're thinking about in your team, it's worth a listen.
What we covered
The episode runs roughly in four parts. Timestamps are from the episode's chapter markers, so you can skip to the bit you are most interested in.
Getting teams experimenting (00:00–08:41)
How AI has changed in a short space of time, and how teams can make room to learn it properly. We talked about running a dedicated AI week for experimentation, and about personal "scratch an itch" projects as a way in. James asked me about mine: Meeting Exec, an executive assistant for cross-company scheduling.
- 01:03 — The evolution of AI and its impact
- 03:13 — AI week: experimentation and learning
- 04:39 — Personal projects: scratch an itch (Meeting Exec)
- 07:35 — Encouraging innovation and creativity
Tech debt, documentation and code review (08:41–19:10)
The core of the conversation: using AI on the work that usually never makes the cut. Paying down tech debt, writing documentation for code that has none, and where AI fits into code review. We also got into where the bottlenecks in development sit once AI is part of the workflow.
- 08:41 — The excitement of AI and its challenges
- 11:19 — Using AI for tech debt and documentation
- 13:46 — AI's role in code review and documentation
- 17:05 — Navigating bottlenecks in development
Maintenance and legacy upgrades (19:10–26:06)
The broader responsibility engineers have for the systems they look after, why maintenance matters, and using AI on legacy upgrades. This is the same territory as my essay: documentation where there was none, tests so you can change things with confidence, then the upgrade itself.
- 19:10 — The broader responsibility of engineers
- 20:26 — The importance of maintenance in software development
- 23:00 — Leveraging AI for legacy code upgrades
Beyond engineering (26:06–end)
AI outside the engineering team: across business functions, in user-facing features for sports technology at Pitchero, and in giving teams the tools to use it themselves. We finished on where AI pricing and development might be heading.
- 26:06 — AI integration across business functions
- 28:59 — User-centric AI features in sports tech
- 31:59 — Empowering teams with AI tools
- 34:47 — The future of AI pricing and development
Listen
You can find it on the Off Script episode page, or wherever you prefer to listen:
Thanks to Josh and James for having me on.