As somebody who likes and uses JetBrains IDEs, I find this fragmentation of their AI strategy (if they have one) frustrating. They now have both Air and ThinkRail in a feature-incomplete state, while the existing AI Assistant – for me far more useful than either Air or ThinkRail – seems to be getting neglected. There are numerous harnesses already, with or without GUIs and editing capabilites; good integration with existing IDE functionality is where there are very limited offerings and where JetBrains could differentiate itself.
As a point of comparison, Zed does the agentic side panel overall better (though still not perfectly), but Zed is still by its nature more limited than e.g. PyCharm or CLion or RustRover or Rider.
Don’t forget about Junie either! I guess they’re throwing everything at the wall and hoping some of it sticks.
That said, I bet that’s how eventually we’ll end up with actually good ADEs (and a bunch of dead projects), by taking what works and changing or discarding what doesn’t - for example GitKraken Kepler has a really nice ticketing system integration but also the older versions I’ve tried had vibe slopped UI.
If nothing else, competition should push everyone who is capable to keep up to do better. It is kind of a bummer sometimes, though - I also wanted their Fleet to succeed, but that was before Zed became really good.
Yes, Junie, Air all got mixed up too much, and it was a bad decision to override some important keyboard shortcuts with Air too.
But, they still have a solid IDE, they just need to figure out what they can do without losing that edge.
I am all for experiments, that's where all the innovations comes from. But don't push VSCode bullshit on me. If I wanted to use that shit, I would use NVIM/Vim. But not VSCode.
Which one exactly? I used to love Webstorm and Rider, but there were so many workflow-breaking bugs in each new release of Webstorm that I switched to VS Code. I still use Rider, but I've stopped my All Products subscription, because new releases usually decreased the amount of well-working features (and new features often didn't work at all).
Around 2022 I'd have agreed 100%, but these days not so much.
Just to add another note, Initially I was a bit skeptic about their new UI. But, they have improved it a lot, and it's really good. Even though I hide most of the UI elements in favor of `double shift`
I use only one, IntelliJ IDEA, which has all integrated, except Rider I believe.
It has full Rust, Java, Python, Web stack, things that I work with.
There were few issues here and there occasionally, but last few months I have not faced any issues, and it's been really robust. It gets slow with LLM providers running in the background on my 7 year old laptop though. But on desktop it flies.
Oh yes to that! It is also built in to IDEA, and It's just so good! I looked for other alternatives, but nothing comes close to their database tool. Works so well.
Oh, I've never really tried that one. It probably goes much further than the DB integrations in Webstorm etc., right? Do you have any favorite features that could be interesting to look at?
I love JB, IntelliJ IDEA is just great, but I am using it less and less with agents now.
I see JB Toolbox has "Junie CLI" available for me. Have you tried both that and ThinkRail or Air?
Which would you recommend for someone who is currently mostly working on the Claude App (which also manages worktrees seamlessly, which looks like what ThinkRail does)?
Junie is integrated in the AI Assistant. It's just one of several agents you can use with it (alongside Codex and various ACP plugins). One option on a dropdown menu.
To this day i don't understand the difference between Junie and AI chat. Anyway, I use PyCharm on a fallback license as a read-only tool nowadays, sad but true.
Junie is an agent. The AI Assistant is the UI for agents (typically in a sidebar). You can switch to other agents if you want (such as Codex or Claude Code), and there are actually a ton that you can add with just a few mouse clicks.
The "Install From ACP Registry ..." allow you to use pi, OpenCode, Kilo Code, Cline, Cursor, Copilot (and a ton of other agents that I've never heard of) if you want.
I feel somewhat bad for JetBrains. Their entire world was flipped upside down. My first intro to a coding harness was through their IDE. Not even a month later and I stopped using their platform altogether. After the year is up I won’t be renewing any of my company’s subscriptions.
They have to try to push this stuff out to survive but none of it is innovative. Not sure why anyone would pay for this when it can be one-shot with a single good prompt. Sure, it looks fancy but I never cared about that.
I can't imagine JetBrains keep existing in the following years. Every programmer I talked to in the past year (myself included) said their job was mostly reviewing generated code. I can imagine why people still use IDE to review code, but I can't name a reason to use something as heavy as JetBrains products just to review things written by others, AI or not.
I have been a paying user for about a decade. Ironically, when I recommended their products to vim/emacs users, they usually were like "thanks, but I know vim/emacs will always exist", and now I realize they really had a point.
You always review large code diffs by comparing them side-by-side? You never need to drop into an IDE to follow the path a branch takes? How do you review large diffs without an IDE, do you remember what every function does in a million-loc codebase?
But vscode and vim do this just as well (in my experience). JetBrains products' advantages over the most lightweight solutions for me are:
- Faster intellisense at least for C# and Java
- Faster syntax errors and other linting-warnings
- Refactor tools
- Debugger and stack state viewer
I still use Rider, but I just don't think faster intellisense/linting matters to me nowadays. Refactor tools feel like something that should come with harnesses. The debugger still helps, but to be honest LLMs are better at debugging than me if you let them dump whatever information they need.
Plus, even with LLMs' help, making and managing a plugin for Rider is a much heavier process than the counterparts for vim/vscode.
IntelliJ is excellent for doing the code review. The git and diff is top notch.
Then there is the DataGrip stuff, I use it all the time.
And the HTTP client (.http files!) is also good. It has built in scripting.
Notice here that none of these activities is writing code.
This enhances the understanding of what I am building with the agents. I’m sure the lazy dev will just skip this and will never see any of this, just let Claude completely take over.
Interesting that you say that, the VCS / diff UI of Jetbrains has always been my favourite by far of any git UI I have tried, I feel like they have an opportunity to capitalise on that a bit more.
For a while I agreed with you, but then I realized I want my agent to have access to the same find references, index, and rename tools, and the rest of the refactor tools that JetBrains have always provided me. And I wanted to be able to use those, even if it means I'll have to pay for Jetbrains.
Just like it was inefficient for me to manually rename variables and classes, it's token inefficient for my agent to do that.
Yeah! Refactoring, is already a solved problem, and deterministic with IDE (unless you have an untyped Python codebase), now we are reinventing the wheel, in more broken ways?
Want to add an additional parameter without creating a thousands of error? Change signature, Extract a part of code to another function? Extract Method.
People are taking these small, but extremely useful conveniences, without waiting, without paying for tokens, for granted.
You can pay a corporate that steals other's hard work without any consequences, $20/m at least. But can't pay another one that is actually putting up a respectable product for so many years, and you at least own it, even if you decided to stop paying?
I have been a paid user for 12+ years. Things got difficult more me this year, but I still renewed, because it's that important a tool for me. And, probably, this is just the hill I die on, owning something, that can't be whisked away at someone else's whim.
Obviously, they are still a corporate, but, until now they've respected me as a consumer unlike many other corporations. I value that.
They have their issues. But, which application does not?!
Their Git merge UI, is just top-notch. And keeps on improving
I might use another coding agent, once JetBrain's credit runs out, but I review, and make all minor changes through the IDE
Project management is just a bliss.
Working with multi-language project? No IDE can manage this how good JetBrains does.
VIM binding is just so good, I don't miss VIM editor.
And the most important point is, they do not push something I do not want continuously. Don't want AI, turn it off, and it's not even there. My data is not being used to train some big LLM, or I am not counted as a product, is really important too.
Yeah, I hope they survive this messy shit, with the core that I have respect for. I hope they don't ruin that.
It's not nostalgia/pity. It's something I have use for.
And, also, it's kind of having a principle, my own. I have issues with corruption manifested by all of these AI corporations. It's an itch that I can't scratch, so I am trying my best to stay away from them.
It's not JetBrains is good and all. But, they're not worse, yet.
Oh look another account created only 12 minutes old doom posting. Getting tired of the unhinged and provocative commentary written by people or bots with ulterior motives.
Why feel bad about it though? If they're unable to innovate and they're stuck in their old ways, and they cannot find/advertise the benefits of either paths by using their products, why would it need to still be around? Isn't this the reason capitalism exist in the first place, the good get rewarded and the bad eventually fold?
I agree with what you are saying, but I'm not sure how much of if applies to JetBrains. They were on the beneficiary side of the same system that flipped their world around for the longest time. I don't feel that bad for them.
Sure, I'd lean more anarchist than anything else on the political spectrum, what exactly am I supposed to be upset/sad about here? Yes, they did great for two decades, so what? So did Blockbuster or Radioshack, but eventually the tides turn and the world changes, this is essentially what life is and always been.
I don't necessarily feel bad about it, but I do feel sad about it.
I would not want all my bread to have come from wheat being milled on an old watermill, but I love visiting those as museums, and I think it would be bad if they were torn down.
I probably feel about the same for JetBrains. Plus, I have fond memories of using them.
I feel bad for them because they seem like some of the better actors in the dev tool space. They make quality products and seem to care about users a lot.
I don’t know why I should feel less empathy for them “because capitalism.” Capitalism has a lot of innocent victims, and in the AI era, there’s no longer something new to move on to.
Oh, I would like to disagree. Specs are transitional and have low value in the long term. The project/product documentation which one should make based on the specs and the actual implementation and found gotcha's - are the truth.
I really don't understand how this is so hard for JetBrains.
Build a polyglot IDE that blends the agentic workflow (terminal chat sessions) with the power of an IDE. Index specs and code so that I can jump from specs to implementation and vice versa. Keep indexing all problems found in the project, with a simple "Ask agent to fix" dialog. Expose debuggers with Markdown output to agents via skills so that they stop trying to debug via logging. Keep the nice UI for build/debug/run configurations, which are still hugely important for reviewing PRs and acceptance testing what the agents are giving you. Upsell by charging to host JetBrains-cloud-isolated-agent-workspaces, like what it looks like Herdr is going to do, so that you're not limited by what you can develop solely on your machine, and development continues after you close the lid, and you can review the output with your teammates in multiplayer, like where Zed is going.
Is it really that hard?? It seems so obvious to me.
I would like to use a harness that integerates with an IDE very easily. This will be helpful for me when coding in a language I largely don't know. I can kinda check things quickly in front of me. I know it just needs an editor and maybe a language server setup – but the cost of visual context switching is still costly heavy for us humans. Often things out of sight are also out of mind.
I was aiming for VS Code but since I have used Intellij and android studio a lot maybe I'll give it a try.
JetBrains should focus on the artisanal engineers and the 'vibing' engineers who need to code by hand from time to time. They have a big moat already. I imagine they try to keep the user base from fleeing their platform, but I think they have already covered prompt integration. There is imho no value in yet another LLM wrapper that seems to be totally unrelated to the existing offering, at least no monetary value.
I would urge JetBrains to put all their LLM power and human brains at work to slash their IDEs backlog and deliver the GOAT of all IDEs. Building IDEs is their strength and there will always be a market for that category.
Pretty sure following this advice would kill them really fast. Only 5% of code will ever need to be hand typed if that. They have analytics that show their IDEs were used by people who deeply card about efficiency and there US nothing more efficient than managing cloud agents.
In a few years we won't even be reviewing code. We will have abstracted code into building blocks and review those abstractions only.
> Only 5% of code will ever need to be hand typed if that.
You are talking about the 'vibing' demografic. Not all of them would use JetBrains, as not all programmers did before the advent of LLMs. But for the time they do have to be in the code, JetBrains could be their best option. They are in a unique position to cover their existing offerings in a gold finish.
> Pretty sure following this advice would kill them really fast.
I don't think so. There is no money to be made in llm wrappers. Their market might have shrunk, sure, but on the other hand they should be able to bring down their expenses significantly while at the same time bringing their backlog to zero. If not, they have nothing to fear from AI and the hype will be over some day.
They're fighting an uphill battle and it's a race against time. I think they've already lost to Claude Code. I can do everything they offer with Claude. Where is their moat?
I was at a large tech conference this week in Amsterdam, it was < 1km from the JetBrains office. They set up the booth 3 hours too late and there was no one actually there when I walked by (twice). I think they have given up?
Im a jetbrains user, dev for 20+ years and Ive never been to a tech conference, I only go to conferences where my customers are. Who needs conferences when you have HN?
A year ago, it was not sure if the core companies around programming would move to more agent enabled workflow. Companies like JetBrains.
I love their products. But it's clear now that programming is now spec management and how do we control agents. I don't need JetBrains for that.
And spec management will become less technical as models get better. What does JetBrains think is the long term goal for this product? I don't think it'll make any sense in a few months.
As usual when there's a harness release by a billion-dollar company, I like to ride its coat-tails with a little plug for mine (https://github.com/juggler-ai/juggler) since my marketing budget is $0. Juggler usually gets a warm response from HN people because it's a bit more hands-on than most of them.
I'd be interested to hear whether the 'spec' stuff works or not. I've been pondering adding a plugin for juggler based on the fact that we're rubbish at writing specs: so instead, when you prompt it, it'd get the model to list the assumptions that it's making about the task. Then the user can spot and correct any incorrect assumptions, rather than expecting the user to provide correct specs. (Haven't actually tried this yet, just speculating)
AS Jetbrains user, tf is this shit, I just want AI integrated better in workflow I already use. If I wanted to use VSCode wrapper I wouldn't pay them for it...
Their plan is so misguided. First they wanted to pay us for separate tool and THEIR AI, which is worse than what people already pay for, few months back they finally figured out nobody wants that and just integrated it into mainline editor and now this
Which is fine if product is good. But they have actually good IDE, instead they decided to make some kind of VS code wrapper and hope people pay it over other editors doing same thing for free ?
Nope. The agent doesn't actually read, and never will. That is, it may pull it into context randomly, and then purge it from context randomly. But any tool/harness/wrapper that claims that agents actually do something is selling you a bridge.
---
Also, the entire page somehow manages present a condensed example of the worst marketing + the worst LLM-adjacent speech in a few short paragraphs.
Of course it depends on the type of software you’re writing and just how strict it needs to be, but in general, I find that you need the “spec-driven development” paradigm less now than you did maybe 6 months ago. Older specs really begin to become noise and context pollutants, plus the model is just going to explore the codebase to get a current state of truth regardless.
As a point of comparison, Zed does the agentic side panel overall better (though still not perfectly), but Zed is still by its nature more limited than e.g. PyCharm or CLion or RustRover or Rider.
That said, I bet that’s how eventually we’ll end up with actually good ADEs (and a bunch of dead projects), by taking what works and changing or discarding what doesn’t - for example GitKraken Kepler has a really nice ticketing system integration but also the older versions I’ve tried had vibe slopped UI.
If nothing else, competition should push everyone who is capable to keep up to do better. It is kind of a bummer sometimes, though - I also wanted their Fleet to succeed, but that was before Zed became really good.
But, they still have a solid IDE, they just need to figure out what they can do without losing that edge.
I am all for experiments, that's where all the innovations comes from. But don't push VSCode bullshit on me. If I wanted to use that shit, I would use NVIM/Vim. But not VSCode.
Which one exactly? I used to love Webstorm and Rider, but there were so many workflow-breaking bugs in each new release of Webstorm that I switched to VS Code. I still use Rider, but I've stopped my All Products subscription, because new releases usually decreased the amount of well-working features (and new features often didn't work at all).
Around 2022 I'd have agreed 100%, but these days not so much.
It has full Rust, Java, Python, Web stack, things that I work with.
There were few issues here and there occasionally, but last few months I have not faced any issues, and it's been really robust. It gets slow with LLM providers running in the background on my 7 year old laptop though. But on desktop it flies.
The have also the Junie [0] alongside their "AI Chat" within the Jebrains IDEs.
[0]: https://junie.jetbrains.com/
Example: https://gist.github.com/rbehrends/4d9febe5c5c008961b309dd4a6...
The "Install From ACP Registry ..." allow you to use pi, OpenCode, Kilo Code, Cline, Cursor, Copilot (and a ton of other agents that I've never heard of) if you want.
They have to try to push this stuff out to survive but none of it is innovative. Not sure why anyone would pay for this when it can be one-shot with a single good prompt. Sure, it looks fancy but I never cared about that.
I have been a paying user for about a decade. Ironically, when I recommended their products to vim/emacs users, they usually were like "thanks, but I know vim/emacs will always exist", and now I realize they really had a point.
But vscode and vim do this just as well (in my experience). JetBrains products' advantages over the most lightweight solutions for me are:
- Faster intellisense at least for C# and Java
- Faster syntax errors and other linting-warnings
- Refactor tools
- Debugger and stack state viewer
I still use Rider, but I just don't think faster intellisense/linting matters to me nowadays. Refactor tools feel like something that should come with harnesses. The debugger still helps, but to be honest LLMs are better at debugging than me if you let them dump whatever information they need.
Plus, even with LLMs' help, making and managing a plugin for Rider is a much heavier process than the counterparts for vim/vscode.
Then there is the DataGrip stuff, I use it all the time.
And the HTTP client (.http files!) is also good. It has built in scripting.
Notice here that none of these activities is writing code.
This enhances the understanding of what I am building with the agents. I’m sure the lazy dev will just skip this and will never see any of this, just let Claude completely take over.
Just like it was inefficient for me to manually rename variables and classes, it's token inefficient for my agent to do that.
Want to add an additional parameter without creating a thousands of error? Change signature, Extract a part of code to another function? Extract Method.
People are taking these small, but extremely useful conveniences, without waiting, without paying for tokens, for granted.
I have been a paid user for 12+ years. Things got difficult more me this year, but I still renewed, because it's that important a tool for me. And, probably, this is just the hill I die on, owning something, that can't be whisked away at someone else's whim.
Obviously, they are still a corporate, but, until now they've respected me as a consumer unlike many other corporations. I value that.
They have their issues. But, which application does not?!
Their Git merge UI, is just top-notch. And keeps on improving I might use another coding agent, once JetBrain's credit runs out, but I review, and make all minor changes through the IDE
Project management is just a bliss.
Working with multi-language project? No IDE can manage this how good JetBrains does.
VIM binding is just so good, I don't miss VIM editor.
And the most important point is, they do not push something I do not want continuously. Don't want AI, turn it off, and it's not even there. My data is not being used to train some big LLM, or I am not counted as a product, is really important too.
Yeah, I hope they survive this messy shit, with the core that I have respect for. I hope they don't ruin that.
And, also, it's kind of having a principle, my own. I have issues with corruption manifested by all of these AI corporations. It's an itch that I can't scratch, so I am trying my best to stay away from them.
It's not JetBrains is good and all. But, they're not worse, yet.
I don't, since I read their privacy policy[0] before deciding to never use their software. Might I add: VSCodium[1] exists.
[0] https://www.jetbrains.com/legal/docs/privacy/privacy/
[1] https://vscodium.com/
I can feel bad about how things went for them, without thinking we are obligated to give them government bailout or something.
I would not want all my bread to have come from wheat being milled on an old watermill, but I love visiting those as museums, and I think it would be bad if they were torn down.
I probably feel about the same for JetBrains. Plus, I have fond memories of using them.
I don’t know why I should feel less empathy for them “because capitalism.” Capitalism has a lot of innocent victims, and in the AI era, there’s no longer something new to move on to.
Oh, I would like to disagree. Specs are transitional and have low value in the long term. The project/product documentation which one should make based on the specs and the actual implementation and found gotcha's - are the truth.
Build a polyglot IDE that blends the agentic workflow (terminal chat sessions) with the power of an IDE. Index specs and code so that I can jump from specs to implementation and vice versa. Keep indexing all problems found in the project, with a simple "Ask agent to fix" dialog. Expose debuggers with Markdown output to agents via skills so that they stop trying to debug via logging. Keep the nice UI for build/debug/run configurations, which are still hugely important for reviewing PRs and acceptance testing what the agents are giving you. Upsell by charging to host JetBrains-cloud-isolated-agent-workspaces, like what it looks like Herdr is going to do, so that you're not limited by what you can develop solely on your machine, and development continues after you close the lid, and you can review the output with your teammates in multiplayer, like where Zed is going.
Is it really that hard?? It seems so obvious to me.
I was aiming for VS Code but since I have used Intellij and android studio a lot maybe I'll give it a try.
I would urge JetBrains to put all their LLM power and human brains at work to slash their IDEs backlog and deliver the GOAT of all IDEs. Building IDEs is their strength and there will always be a market for that category.
In a few years we won't even be reviewing code. We will have abstracted code into building blocks and review those abstractions only.
I love their products. But it's clear now that programming is now spec management and how do we control agents. I don't need JetBrains for that.
And spec management will become less technical as models get better. What does JetBrains think is the long term goal for this product? I don't think it'll make any sense in a few months.
I'd be interested to hear whether the 'spec' stuff works or not. I've been pondering adding a plugin for juggler based on the fact that we're rubbish at writing specs: so instead, when you prompt it, it'd get the model to list the assumptions that it's making about the task. Then the user can spot and correct any incorrect assumptions, rather than expecting the user to provide correct specs. (Haven't actually tried this yet, just speculating)
Their plan is so misguided. First they wanted to pay us for separate tool and THEIR AI, which is worse than what people already pay for, few months back they finally figured out nobody wants that and just integrated it into mainline editor and now this
An AI slop webpage trying to sell me VS Code bullshit. Really?
Nope. The agent doesn't actually read, and never will. That is, it may pull it into context randomly, and then purge it from context randomly. But any tool/harness/wrapper that claims that agents actually do something is selling you a bridge.
---
Also, the entire page somehow manages present a condensed example of the worst marketing + the worst LLM-adjacent speech in a few short paragraphs.