Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Google fixed more Chrome bugs in June than over the past two years, thanks to AI

blog.google

Post reply on HN
Print view View original

351–360 of 668 posts

Re: Google fixed more Chrome bugs in June than over the past two years, thanks to AI

#351

Post by dragonwriter »

dominotw wrote: post #244
herrkanin wrote: post #34

Earlier quoted context omitted.

The thing that makes it work really well is to make sure it has all the tooling to verify its hypotheses. If you allow it to run the full lifecycle in loops you will be surprised how well it works.

Very few problems really have "all the tooling to verify its hypotheses" though. even if you want to construct such an harness. Also let me ask you why we need better and better and models if what we have already can produce good output with 'all the tooling to verify its hypotheses'

> Also let me ask you why we need better and better and models if what we have already can produce good output with 'all the tooling to verify its hypotheses'

“Good” isn’t “perfect” and even if it was, the ability to produce perfect output with all the tooling to verify its hypotheses could still be improved, in time and token efficiency, by better models producing fewer spurious hypotheses, rejecting those it does generate faster, and taking fewer unnecessary steps in confirming its good hypotheses.

Re: Google fixed more Chrome bugs in June than over the past two years, thanks to AI

#352

Post by andrehacker »

glimshe wrote: post #41

A lot of people here seem to be living in a different universe than me or simply don't know how to work with AI. I think detractors believe you should just let AI do the job blindly instead of leveraging it as a tool to accelerate you. They get mad at Excel for the poor investment returns. At this point, this is such a strawman, it isn't worth counter arguing. I think I'll abandon this discussion and keep using AI qu…

It appears to me that this is the same pattern you often see when junior employees are given one or more resources intended to help them succeed. Despite having access to those resources, there is frequently no shortage of complaints about how incompetent people are, that they "should know" esoteric home-built sowtware patterns or business knowledge without being told or that they failed to perform a task in the one specific way that someone considers the "correct" approach.

Re: Google fixed more Chrome bugs in June than over the past two years, thanks to AI

#353

Post by Leynos »

Tade0 wrote: post #86
madeofpalk wrote: post #72

Earlier quoted context omitted.

Tests! Unit tests, integration tests, random adhoc scripts. You know - TDD! I’ve been working on UI component improvements and it was doing a lousy job until i specifically told it to test in a headless browser to validate it works. I think somewhere in an AGENTS.md i have an instruction to “don’t state your guesses as fact - validate findings and results”.

It bothers me that you have to explicitly state this to the agent. Makes me think what else is missing from that file which also needs to be explicitly stated, but I don't know what don't know. "Do a good job"?

If you ask Fable or 5.6 Sol to improve performance, it will generally know to build a benchmark and create a test corpus. I'm not sure where the contrary suggestion is coming from.

Re: Google fixed more Chrome bugs in June than over the past two years, thanks to AI

#354

Post by 9x39 »

dpark wrote: post #228
cyanydeez wrote: post #156

Earlier quoted context omitted.

the point is more: are people going to flood the zone with infinite bikeshedding that drowns out the real gains

What is bikeshedding supposed to mean here?

It's a low-effort drive-by comment, but I read it as tourists prompting about an issue they don't really know about.

People will recognize a glimmer of something they can attach to, prompt and get a few pages of junk about this narrow topic, and then contribute it in a way that further reduces the signal-noise ratio of the discussion or project.

Like, dude, we have the same tools. We can get the same slop from the tap at any time.

Re: Google fixed more Chrome bugs in June than over the past two years, thanks to AI

#355

Post by majormajor »

stack_framer wrote: post #301
VBprogrammer wrote: post #17

I've recently been using AI a lot for performance optimisation during a particularly busy period at work. I would say it was almost completely useless at the high-level direction - it would point out suspicious parts of SQL queries for example but on back to back testing these almost never resulted in any performance change. In fact, if it wasn't for the fact that it made making the actual changes I identified much e…

The useless suggestions are exhausting. One of my co-workers uses Claude for all asynchronous communication, including Slack messages, Jira comments, code reviews, and emails. Every single message from him, literally every time he communicates, it's a massive wall of text, overflowing with scope-creep suggestions like nothing I've ever seen before. It's impossible to ask him a simple question and get a simple answer.…

What have your or their manager said about this? About any or all of (a) slowing everyone else down with crap, (b) wasting company resources on generating crap, (c) not doing anything that a CI integration couldn't trigger the AI to do itself?

Re: Google fixed more Chrome bugs in June than over the past two years, thanks to AI

#356

Post by majormajor »

harrall wrote: post #326
programmertote wrote: post #317

Earlier quoted context omitted.

I have at least one coworker who literally copy-paste AI output to respond to code and doc reviews (and most of her communications like on Teams and Asana) despite me telling her multiple times to take it a notch down. Two days ago, she started answering my doc review questions with literally AI outputs (like "Good question. " and "You're right. "). That really bothered me that I finally announced AI "code of conduct…

If I were to be real, my coworkers that post AI responses straight rarely had insightful comments before AI.

> If I were to be real, my coworkers that post AI responses straight rarely had insightful comments before AI.

There's never been a shortage of low-value coworkers because hiring is hard and firing is unpleasant and painful in many ways.

But there's a weird phenomenon with AI where a lot of people are using it to actively call attention to the fact that they aren't doing anything but call the tools.

Rarely in the past have the I-just-want-to-coast-doing-the-minimum folks called so much attention to it!

Re: Google fixed more Chrome bugs in June than over the past two years, thanks to AI

#357

Post by bonoboTP »

qsort wrote: post #66
sigmoid10 wrote: post #30

Earlier quoted context omitted.

Which model/agent/harness tool did you use? I've found what you describe was my exact experience some ~6-8 months ago, but since about a month or so the game has completely changed. Using 5.6 Sol with highest reasoning setting in Codex or Fable in Code, the models come up with a list of possible improvements from static analysis (ranked by complexity/benefit), write and run their own custom profilers and deliver sign…

Very hard to say anything definitive on this because it's a moving target, but last time I tried models still had a distinct sense of "consistently good, sometimes great at micro, bad at macro". Similar to how, even for relatively pedestrian CRUD, they'll do code that's objectively fine at the function/file/class level but can still make a mess if you don't supervise them at least at a high-level.

It is moving so fast that your experience with older models is irrelevant today. Sorry.

Claude Code, Fable 5, xhigh reasoning, allow it to run the full CI, end to end and benchmark, it will not make silly mistakes (or only occasionally). Also, be able to state what you desire. Have any docs or materials in the same directory so the model can reference it. For even better results: turn on speech recognition and braindump what you know about the system, its goals, its context, history anything relevant, any gotchas you'd explain to a new employee or intern. Talk for 5-10 minutes. This is optional, "make it faster" can already get a large part of the job done.

And if it doesn't work well, describe what you dislike in its solution and tell it. Even just one extra iteration can make things work.

(I guess GPT-5.6 can be similarly good, I use Claude)

I feel like some people are emotionally invested in it not working and subconsciously sabotage their own effective use of the tool.

Re: Google fixed more Chrome bugs in June than over the past two years, thanks to AI

#358

Post by swat535 »

Tade0 wrote: post #86
madeofpalk wrote: post #72

Earlier quoted context omitted.

Tests! Unit tests, integration tests, random adhoc scripts. You know - TDD! I’ve been working on UI component improvements and it was doing a lousy job until i specifically told it to test in a headless browser to validate it works. I think somewhere in an AGENTS.md i have an instruction to “don’t state your guesses as fact - validate findings and results”.

It bothers me that you have to explicitly state this to the agent. Makes me think what else is missing from that file which also needs to be explicitly stated, but I don't know what don't know. "Do a good job"?

Additionally, if you are asking agents to write test, be very careful. They will write tautology tests.

They will mock things to no end. They will flat out REMOVE assertions (saying it's not needed). They can also write test to assert the wrong result.

You have to always review it, it's exhausting.

Re: Google fixed more Chrome bugs in June than over the past two years, thanks to AI

#359

Post by tracker1 »

rootusrootus wrote: post #305
stack_framer wrote: post #301

Earlier quoted context omitted.

The useless suggestions are exhausting. One of my co-workers uses Claude for all asynchronous communication, including Slack messages, Jira comments, code reviews, and emails. Every single message from him, literally every time he communicates, it's a massive wall of text, overflowing with scope-creep suggestions like nothing I've ever seen before. It's impossible to ask him a simple question and get a simple answer.…

As much as I don't truly love the shop I'm at now, stories like this keep me from looking for anything new. At least until the AI psychosis [hopefully] passes. I can't even get my management to approve a $100 Claude Max subscription, which is kind of annoying but also means they have not changed their expectations on what the team produces. None of my coworkers is going to be shoveling AI slop my direction because th…

Yeah, I can't even run AI on work hardware... I have been able to use it for writing small utils/libraries that I then pull into the work... but the divide is clear and I review all the code myself.

For a couple examples, working through an animated loader for html/js/css with an svg for the org. Another was working through a library implementation to work against an interface that was designed for Mongo, but the org is using SQL Server. Latest was a quick util to extract a zip file of pdfs into a 1bit(b/w), zopfli compressed png file per page.

Generally stuff I could do, but would take me a few days for research and experimentation vs an hour or two with AI.

Re: Google fixed more Chrome bugs in June than over the past two years, thanks to AI

#360

Post by majormajor »

jfrbfbreudh wrote: post #83
herrkanin wrote: post #34

Earlier quoted context omitted.

The thing that makes it work really well is to make sure it has all the tooling to verify its hypotheses. If you allow it to run the full lifecycle in loops you will be surprised how well it works.

Built exactly this for work a few months ago. Claude has a read-only account to the dev DB and a bunch of python postgres tools, I point it at queries/views and ask it optimize, and it has to: 1) verify identical query results 2) run repeatedly to get average, worst, best, etc duration of runs Sped up so many legacy things that none of us were ever going to bother with.

What if the shape of data in the dev db doesn't stay consistent with real life?

Is the harness for the AI set up with any sort of instructions to generally prefer broadly-applicable first-principals query performance analysis over unintuitive results that may be local maxima due to certain things specific to that env?

Even with humans I've had to unwind "optimizations" before that worked great for low-volume envs by taking advantage of 'be inefficient in the small scale with memory to save CPU and wall clock time' or such in a way that caused pain deployed.

Post reply on HN