krzyk

I did similarly with copilot.

I have about 15 or so subagents doing reviews from different perspectives (or providing some additional value, like finding agents.md files, doing confidence ranking, describing images attached to the PR, that get validated later on with Jira issue description).

I used it since about November, with large scale popularity in my company reaching in April - all that on a 300 premium requests (because they allowed starting subagents, and there was no limit how long a single request can last) - so it would cost something like $5000 and $8000 for April and May if it was API pricing. I had similar cost per review (about $0.90) with Opus 4.6 and help from Sonnet and Haiku for simpler tasks. It did about 4000 reviews during the last 2 months.

And starting in June, it will be dead because it will be API pricing and for $30 (or $19 since September) it will do just few reviews.

A fun project.

It is starting the review during CI (CI just triggers the review), not blocking merges like failed build or lint failures.

Actually not, it is similar debate like rebase or merge.

e.g. I don't squash, I prefer to see full history, not redacted one.

It is easier to view code review results in a tool and not in a text during commit.

There is no universal standard which IDEs support for code review results (there is SARIF, but it is not supported that widely). A review result on a web page with comments from humans, is valuable.

codex is also good, has better usage limits compared to CC.

Issue is that CC forced corps over 150 people into a API pricing, which is, well, suboptimal compared what we get. I think it will push those towards hiring more juniors (finally).

What is strange, that medical matters are mostly translated to Polish, while when I watch medical shows on TV I see that US still tends to use Latin there.

Similar thing for species names, not sure where I've seen it.

A side note from Polanmd: My professor of Solids State Physics was pretty fluent in both Latin and Greek, which I find quite interesting (it was about ~20 years ago, he was in his 60-70s).

Why carrying it is practical? What is it used for?

The only time I need my ID is during elections, but I can also show the one I have in our government produced app.

Older people in Poland do carry those, mostly out of some kind of habit and some kind of fear that police might need it. I can drive a car and get a speeding ticket, and all I need police to know is my ID number (it is not the identifier of my identity card), which I know by hard, it consists of my birth date plus 5 semi-random numbers). I don't need my ID, my driving license nor my insurance data - everything is located in police database based on my ID number (or my license plate).

Bad sides of carrying it is that you might loose it, and that is a PITA, because you need to block it right away (someone might take a loan with it, happens I kid you not) by calling your bank or similar service.

So I take it out from my wallet (which I don't carry also) only when I go to the airport.

Maybe better, but less useful. I don't carry my Identity Card at all, unless I cross the border within EU where it is used. All other functions I have in our country app. To which I can log in using physical card, but I have other options that are online.

Some see it as a promotion other (like me) as a demotion. I still prefer to do it myself, although I like code reviews done by AI, they do help to make code a bit better.

I had similar experiences (post-communist country), but I was a bit older, got my first computer at age of 9 (Atari 65XE, replaced with C64 2 years later).

So I got a bit of childhood before receiving those (pretty scary stuff if I think about it - exploring holes that looked like trenches, that were dig up by workers to put water pipes in them, or exploring an old, ruined house) - living in a typical communist-era block of flats community.

But a year or so after receiving computer I started spending less and less time outside and more on computers, basically making me socially isolated except my two closest colleagues. Nowadays my two daughters are more social than I am, but I like computers too much.

You do know it is not literal transformation?

It would be great if they didn't name things to similar things that already exist. Raspberry Pi is quite popular and I think it should be known for the author.

Opencode (and other coding agents) have hundreds of open issues reported. It is quite discouraging when they are not being closed/fixed.

That name is quite near the infamous Electron, is it similar?

Now as you can see from the article, it starts turning. People are getting less pricey than agents on API pricing.

Copilot switches to API pricing starting next month (let's see how long it will last for our $39, and $19 since September), Anthropic switches all corps into API based pricing. From the most popular choices I think only Codex didn't switch yet (although it is hard to tell because I don't know their enterprise pricing).

Oh, I didn't know about that. We got into Claude Code in AFAIR December, and we were on API pricing from the beginning.

Any corp (> 150 seats) has to use API pricing, so e.g. I don't have pressure or weekly limits, just a set budget I can use each month.

But you have IDE for you and cli for agent. Agent works on the same code, you can see the changes right there.

Our corp specifically prohibits that, because of code leak/training.

In my model I have also opis 4.6.

Maybe this is becaus I'm on api pricing? (All new contracts for corps are pushed to that by Anthropic).

Harness makes a difference. Also in copilot you have smaller context for Claude models.

And you get a token based pricing since June 1.

If you have such toxic environment, run.

Maybe the GP was more about: "Why have agent in IDE?" kind of comment, which I understand. Having a small window into agent is annoying for me. I prefer to have coding agent in CLI and do my editing in IDE (and seeing the code changes there, I still see no point in CLI agents providing me with diffs, I can see those using e.g. git diff or IDE).

> Why is Google the 2nd or 3rd tier player in the AI coding market?

2nd, 3rd? No way. You either use Claude Code or Codex, the 3rd option usually was Github Copilot. The only time I heard someone used google for code writing it was Linux Torvalds in one of his commits.

I never thought about that, good idea.

Isn't CUDA/nvidia the go to solution for most local models, with the rest being second class citizents?

It is sad. I like programming, if I couldn't do it and had to write text (which I do hate, I'm not a writer) it would be make quite a sad world.

People in my company sounded underwhelmed by it. It usually was founding issues by not understanding deployment (or not being fed that info).

Yes.

There is also homebrew install.

What is the use case for integrating coding agent in IDE?

I use run agents outside of my IDE, while they work I can look at the code they created, or I can us IDE to do different work.