Continuous Delivery in a World of Constant Change • Abby Bangser & Dave Farley • GOTO 2025

This presentation was recorded at GOTO Copenhagen 2025. #GOTOcon #GOTOcph https://gotocph.com Abby Bangser - Principal Engineer at Syntasso & Team Topologies Advocate @SyntassoAbby Dave Farley - Bestselling Author, Founder & Director of Continuous Delivery Ltd. @ModernSoftwareEngineeringYT RESOURCES Abby https://bsky.app/profile/abangser.bsky.social https://twitter.com/a_bangser https://github.com/abangser https://www.linkedin.com/in/abbybangser https://www.syntasso.io/members-area/abby/profile Dave https://bsky.app/profile/davefarley77.bsky.social https://www.continuous-delivery.co.uk https://linkedin.com/in/dave-farley-a67927 https://twitter.com/davefarley77 http://www.davefarley.net DESCRIPTION Dave Farley and Abby Bangser open with a clear statement: Continuous Delivery isn't a relic of the pre-AI era — it's the foundation that makes the AI era survivable. Dave's definition is simple but consequential: software should always be in a releasable state, verified after every small change. That's not just a workflow preference; it's the same incremental, hypothesis-driven approach that underpins science and engineering. In an AI-assisted world where code can be generated far faster than humans can reason about it, the discipline of small, safe, verifiable steps becomes more critical, not less. The danger isn't AI writing bad code — it's AI writing a lot of code very fast that nobody is properly checking. The conversation turns to a genuinely alarming DORA report statistic: 70% of developers using AI tools don't distrust the output. Abby draws a parallel to the long-running debate over whether developers can be trusted to test their own code — they usually can't, without a deliberate change in perspective. The same challenge applies to AI-generated code: you need to consciously shift from "prompter" mode to "verifier" mode, and most developers aren't making that switch. Dave closes with a surprising note of optimism: AI may be the industry's best-ever opportunity to finally get XP practices — small increments, automated tests, continuous feedback — embedded into how teams actually work. Not because anyone chose to adopt them ideologically, but because working without them while using AI is visibly, measurably risky. TIMECODES 00:00 Intro 01:45 Is CD still relevant in an AI world? 06:38 What does CD look like when AI writes the code? 21:27 Is SW development a solved problem? 39:12 How do you stop AI from cheating on its tests? 45:01 Outro Read the full abstract here: https://gotocph.com/2025/sessions/3779 RECOMMENDED BOOKS Kief Morris • Infrastructure as Code • https://amzn.to/4e6EBQc Matthew Skelton & Manuel Pais • Team Topologies • http://amzn.to/3sVLyLQ Dave Thomas • simplicity • https://amzn.to/43FghBJ Dave Farley & Jez Humble • Continuous Delivery • https://amzn.to/3ocIHwd David Farley • Modern Software Engineering • https://amzn.to/3GI468M Dave Farley • Continuous Delivery Pipelines • https://amzn.to/3rjetdi https://bsky.app/profile/gotocon.com https://twitter.com/GOTOcon https://www.linkedin.com/company/goto- https://www.instagram.com/goto_con https://www.facebook.com/GOTOConferences #ContinuousDelivery #ContinuousArchitecture #EvolutionaryArchitecture #PlatformEngineering #PlatformAsAProduct #PlatformTeams #Platforms #Observability #FeedbackLoops #TodayInTech #Programming #SoftwareEngineering #AbbyBangser #DaveFarley CHANNEL MEMBERSHIP BONUS Join this channel to get early access to videos & other perks: https://www.youtube.com/channel/UCs_tLP3AiwYKwdUHpltJPuA/join Looking for a unique learning experience? Attend the next GOTO conference near you! Get your ticket at https://gotopia.tech Sign up for updates and specials at https://gotopia.tech/newsletter SUBSCRIBE TO OUR CHANNEL - new videos posted almost daily. https://www.youtube.com/user/GotoConferences/?sub_confirmation=1