Charles Pan

building @stagereviewapp (YC P26)

[Charles Pan @ceefryingpan]

@GaddipatiHarsha 🫡

[Charles Pan @ceefryingpan]

@unclebobmartin rebasing has never been easier 🙏

[Uncle Bob Martin @unclebobmartin]

There is no way to get the human out of the loop. The Agents simply do not understand obvious user expectations.

[Charles Pan @ceefryingpan]

@benhylak Should check out @StageReviewApp

[Charles Pan @ceefryingpan]

@ycombinator 🚀🚀

[Charles Pan @ceefryingpan]

I'm so tired of agents overengineering stuff. Has anyone figured out a robust solution?

[Dean Stratakos @DeanStratakos]

Lightning quick keyboard shortcuts ⚡⌨️

Use `?` to see all shortcuts

[Charles Pan @ceefryingpan]

Can’t believe YC is coming to a close in 2 weeks, feels like we just got started yesterday

Enjoy some of the pics me and @DeanStratakos took so far through the many ups and downs

[Charles Pan @ceefryingpan]

@icanvardar codex

[Charles Pan @ceefryingpan]

@dexhorthy we need to have a serious conversation about model performance plateauing

[Charles Pan @ceefryingpan]

@theo not to mention - Claude Code also feels much slower than it was a few months ago

[Charles Pan @ceefryingpan]

@SzymonRybczak we use Stage to review Stage 🫡

[Charles Pan @ceefryingpan]

why are more people not using `/model opusplan`?

seems like a no brainer to me: you get the problem solving of Opus and the efficiency + speed of Sonnet.

[Charles Pan @ceefryingpan]

one lifesaver feature that I always wish I had in GitHub

toggle ability to hide bot comments

[Charles Pan @ceefryingpan]

@pierrecomputer this is exactly what I was looking for

[Charles Pan @ceefryingpan]

More people need to hear this

[Mitchell Hashimoto @mitchellh]

I've got an agent in a loop optimizing a renderer with the goal to minimize frame times (and tests to measure). It got times down from 88ms to 2ms and allocations down from ~150K to 500. Sounds good, right? Wrong. This is exactly why agent psychosis is a big fucking problem.

As an experiment, I rewrote the Ghostty core render state in Go, with access to identically laid out data structures as Ghostty and the exact same validation tests. I made a purposely naive renderer (simple, correct, but slow). 88ms per frame with 150,000 allocations (horrendous, lol)!

I then kickstarted a Ralph loop to bring the frame times down. I told it it can't modify input data structures or the public API or tests (they're correct), but it can do anything else it wants. It got to work.

It has worked for about 4 hours. I've spent around $350 on this experiment so far. The results?

88ms => 1.5ms
150K allocs => ~500 allocs

Incredible right? Nope.

My hand-written renderer I ported has frame times (same benchmark) of ~20us (0.020ms) and 0 allocations in the update path.

This is the problem with psychosis and lacking systems understanding. If you don't understand the system, you're going to accept that this is an incredible result. If you understand the system, you'll see better solutions immediately and can do roughly 75x better on throughput.

The people who blindly trust agent output are in the former camp. They're sheeple, overdrinking from a fountain of mediocrity.

Standard disclaimer: I use AI all the time. I like AI. The point I'm making is to not blindly accept results. Think. Analyze. Learn.

[Charles Pan @ceefryingpan]

we kept hearing from people that they only focus on high risk areas when reviewing PRs

so we made that way easier on @StageReviewApp

[Charles Pan @ceefryingpan]

@odysseus0z @Liggi @steipete Retired now!

[Charles Pan @ceefryingpan]

@FlyaKiet @superset_sh This is huge

[Charles Pan @ceefryingpan]

@odysseus0z @Liggi @steipete great person too, I got to take a class with him a few years ago