CG
Chris Gmyr
X (Twitter) ↗ LinkedIn ↗ Mastodon ↗ Bluesky ↗ Instagram ↗

Chris Gmyr

Things Chris Says on Podcasts

Staff software engineer working on Laravel and PHP, currently at Rula and a co-organizer of TrianglePHP. Hosts Slightly Caffeinated, a podcast about Laravel development, AI coding tools and side projects, covering topics like ClickHouse migrations, Prism (a Laravel package he works on) and coffee brewing. Co-hosts the show with TJ Miller and has also appeared as a guest on North Meets South Web Podcast.

Where to Find Them

Chris Gmyr hosts and writes Slightly Caffeinated . They have also been a guest on North Meets South Web Podcast (7 times) .

Recently: “Chemex, Prism's Comeback, and Avocet Goes Live” on Slightly Caffeinated (October 2026); “Fall Vibes, AI Evals, and Code Review Bots” on Slightly Caffeinated (October 2026); “New Models, Open Source Hive, and Avocet” on Slightly Caffeinated (September 2026); “Peru Beans, AI Doom, and Code Review Burden” on Slightly Caffeinated (September 2026); “Laptop Rescue, Rig, and Decision Fatigue” on Slightly Caffeinated (September 2026); “Starday Coffee, Book Club App, and Tauri vs Swift” on Slightly Caffeinated (September 2026).

What They Said

“Adversarial by nature is adversarial. It's built to find something no matter what. … And then now this little itty bitty change has exploded into five, 10 different commits … your scope has blown up 10x, your PR has blown up 10x. … Why can't we just ship it?” — Chris Gmyr, Slightly Caffeinated

Gmyr runs adversarial AI reviews on his own projects. Each round finds one more edge case, the fix invites another round, and a small change ends up handling 20 cases nobody asked for. He and TJ Miller are still working out where to draw the line.

Slightly Caffeinated · 2026-09-19 Permalink → Listen →
Slightly Caffeinated Around 34:59 into the episode
Chris Gmyr

…CodeRabbit before for personal stuff. I'll have to look at it again because I do the adversarial reviews too. And also I'm trying to build my own set of review skills for my personal projects. But it is so hard to find the balance of like, yes, you found something, but is this actually important?

TJ Miller

Right.

Chris Gmyr

…a valid, you know, call out. Let me fix that. And then the next thing, it bypasses the thing that you just fixed, but now it finds issue number two. Oh, yeah, issue two makes sense. We should fix that. And then now this like little itty bitty change has exploded into five, 10 different commits and it's still going and still being adversarial and finding these like little nitpicky things. And your scope has blown up 10x, your PR has blown up 10x. And now it's just like my one, you know, little change to a component or even, you know, a design or whatever the case is. It doesn't even matter. It's like it's blown completely out of proportion and handles all these like 20 different edge cases. And it's like, why? Why can't we just ship it? And then if an issue comes up, then we'll fix it then. And, you know, some things like it does find…

TJ Miller

Yeah, I was literally just wrestling with that yesterday of like, it just kept on finding things no matter what. And it's like, how do I, how do I prompt its way into, you know, where that line even is? And like, how do you define that? And geez, like, it's, it's, it feels really nuanced. And cause…

Chris Gmyr

that. Yep, exactly. And the other thing that's been on my mind is like how to test and tune these things. So with like my AI review script bot, whatever, I do a lot of like manual testing and I'll run like the dev branch on like repos that I own. So I can be like, you know, how does this feel…

TJ Miller

I, Claude and I built an eval harness as part of the work because that's, that's just like every AI feature that I have built at Luma for the platform. I build eval harnesses to go along with them so so that Claude can iterate on its own and kind of like run the eval, tweak some things, run the…

Speaker names from our own diarization · position estimated from where the line sits in the episode
“If you're going to call something out, validate that you can actually point to the thing that you're calling out. And if you can't, then just drop it. So just that call out increased the correctness of the review probably 15 or 20 points.” — Chris Gmyr, Slightly Caffeinated

Gmyr and TJ Miller are both building in-house AI pull request reviewers at their companies. His bot kept flagging problems that weren't there; making it cite the exact line of code it objected to cut most of the hallucinated findings.

Slightly Caffeinated · 2026-10-10 Permalink → Listen →
Slightly Caffeinated Around 21:11 into the episode
Chris Gmyr

Cool. So last couple of times we talked about a whole bunch of my projects going on, but I wanted to circle back to you and some of the Luma projects that you've had going on recently.

TJ Miller

…examples of like, here's, here's what's worth calling out. Here's not, you know, here's stuff that doesn't matter. Yeah, it's, it's so nuanced. And that's been really, really tricky to nail. Like I said, technically, the code and the architecture and all that, not the challenging piece here at all.

Chris Gmyr

Yeah. Yeah, totally. I know we talked about this a little bit before, but basically we're working on the same project, two different companies. But yeah, so I don't remember how much in-depth we went to into last time. But yeah, for the fixtures, I have a whole nother separate repo outside of the PR reviewer, like GitHub action repo to pull in any of these fixtures. So I went through because I've gotten a bunch of feedback from other people around the company in different tech stacks, different teams. And it's like, hey, this PR is just, you know, the review is just not good. And I made comments and it didn't do anything, which it's expected because we weren't pulling in comments and stuff like that at the time. It's just like a fresh eye review, like knows nothing about anything really besides like the diff that we pushed in it. So I've been able to capture, I don't know, probably 20, 25 specific PRs from across different teams and tech stacks and reads of like review and comments and all that stuff and have a bunch of…

TJ Miller

Yeah, I really, I really like. So what I had done before is I had it do rule matching on the glob patterns, just how like Claude has it. So it was only adding those to context as needed. And I had something in there before of like your review has to be rooted in rules. So like kind of like you have…

Chris Gmyr

Yeah. Yeah, exactly. So yeah, the agent side is moving. And then I've also tried to make some movements on the human side of reviews as well. So I've been making a bunch of changes to our engineering skills. And there's one in there for like PR create. So if you do this, it'll like open it up as a…

TJ Miller

Yeah, we went and made a config file that lives in the root of the repo that has basically glob paths for must-have human review for like auth and like migrations and stuff like that, where it's like that gets called out automatically in the PR, like the agents comment on the PR of like, these…

Speaker names from our own diarization · position estimated from where the line sits in the episode

Collections They Appear In

Hands-On Engineering Podcasts Laravel Ecosystem