12 quotes from 1 episode on The Changelog, each with a timestamped link to the source.
12 quotes1 episode
The short version
Dave Kerr states that deploying AI agents involves rigid operational boundaries and visual proof of success. Visual verification through terminal recordings and audit logs confirms correctness beyond basic unit testing.
Most interesting insights
Regulated clients demand built-in audit logs that track resource creation and editor identity before deploying any AI tools.
“This client need to know which person created which query and who edited it. Can we get this out of the box? And it's like, 'Yeah, we've got audit logging. It will tell you the resources that have been created and who created them.'”
Dave Kerr, The Changelog · November 2026 · Listen ↗
Production automation treats AI agents as governed services within Kubernetes. Clients prefer predictable Python scripts that fan out operations to limit open-ended prompts.
“…one of the things that my team are building and building an open source at the moment, it's a set of like tools for Kubernetes that mean you can run agents, and treat them a little bit like services you'd deploy and you'd manage and govern.”
Dave Kerr, The Changelog · November 2026 · Listen ↗
“…a lot of the time they wanted deterministic steps. They don't want to say to an agent, 'Read this folder and do X.' They want to say instead, with a bit of Python code, 'List the contents of this folder, and then for each one, fan out an operation.'”
Dave Kerr, The Changelog · November 2026 · Listen ↗
Large language models fuel addictive developer grandiosity.
Fast AI responses act as unconditional validators that remove human friction. This loop generates rapid context switching and hyper-creativity that leads to sleepless nights.
“LLMs are very good at feeding your grandiosity and feeding your ideas, 'That's a great idea. What about this?'”
Dave Kerr, The Changelog · November 2026 · Listen ↗
“…very fast context switching, hyper-creativity, this grandiosity of 'I've got this idea and I can turn it into reality.' And that's addictive, and I'll stay up all night doing this.”
Dave Kerr, The Changelog · November 2026 · Listen ↗
Mental recovery takes scheduled time away from screens.
Processing intense creative feelings involves stepping away from work entirely. Dave Kerr religiously shuts down laptop tasks to sit with ideas without immediately solving them.
“The idea of integration is that when you have those thoughts and feelings, you'll need time to just be with them for a while, not necessarily solve them, but just let yourself be there…”
Dave Kerr, The Changelog · November 2026 · Listen ↗
“For whatever the the bug fix is, give me evidence that it works. So, then you get your terminal recordings that you can then look at and eyeball, and if you see that column is there, you're like, 'Okay. That passes the smell test. I'll now start to go into detail.'”
Dave Kerr, The Changelog · November 2026 · Listen ↗
“Don't just write the code. Build a body of evidence that gives me confidence that what is there is correct, whether screenshots, or recordings, or log log entries, or examples of what you did.”
Dave Kerr, The Changelog · November 2026 · Listen ↗
“…evidence is not the unit tests pass, cuz if the unit tests don't pass, I'm canceling my subscription, cuz you know they really should pass. To me, that's not evidence. That's just like, well, that's hygiene that has to be there.”
Dave Kerr, The Changelog · November 2026 · Listen ↗
Dave Kerr built Arc, an open-source toolset that turns AI agents into standard Kubernetes workloads instead of messy local Python scripts.
Regulated enterprise clients demand audit logs, secrets management through tools like HashiCorp Vault, and role-based access control before running any LLM code.
Dave Kerr presented at AI DevDays on how coding agents dysregulate neurodivergent builders by generating unchecked excitement and cascading behaviors.
Large language models act as unconditional yes-men, validating manic grandiosity with replies like "That's a great idea!" while removing the human friction that exposes bad assumptions.
Green unit tests are basic hygiene rather than proof of correctness, because autonomous agents routinely write self-serving tests that pass without solving the underlying defect.
Dave Kerr built Shellwright, an open-source tool that creates pseudo-TTYs to record before-and-after terminal sessions for agent pull requests.
How we attribute quotes. Every quote was matched against the episode transcript, so the words and the timestamp are real (we trim filler words like "um", nothing else). The name comes from our written summary of the episode, and we use it only when a separate check of the captions finds that person on the episode. YouTube gives us no voice-by-voice transcript, so open the timestamp to hear who is talking. See a wrong name? Tell us and we fix or remove it.