Inside JetBrains—the company reshaping how the world writes …

Inside JetBrains—the company reshaping how the world writes …

OpenAI News

如果你不是写软件的,可能不知道 JetBrains ;如果你写软件,几乎肯定在用他们的产品。

这家公司在现代开发的幕后发挥作用——其工具被全球约 1500 万名专业工程师使用(包括 88 家 Fortune 100 ),并且是 Kotlin 的创造者(Android 的官方编程语言)。只要你打开过 IntelliJ 、 PyCharm 、 WebStorm 、 GoLand 或 Rider ,就用过 JetBrains 。

我们与 JetBrains 的产品负责人 Kris Kang 交谈,探讨团队如何把 OpenAI 的模型引入开发流程,不是要取代开发者,而是提高他们的上限。

“开发者不仅仅在写代码。他们还要审查、推理、设计系统。AI 可以帮助处理那些超出单纯敲代码的部分。”—— Kris Kang , JetBrains 产品负责人

JetBrains 如何采用 OpenAI

“超过 1500 万名开发者在用 JetBrains ——现在我们把 OpenAI 带入他们的工作流。” Kang 说。这一转变不仅是自动化,更是赋能:保护开发者的专注流、减少重复劳动,让工程师把精力放在设计、架构和判断上——这些是在 AI 辅助下能长期放大的能力。

在内部, JetBrains 团队使用的模型包括:

  • ChatGPT
  • GPT‑5
  • Codex

对外, JetBrains 的客户可以在编码代理 Junie 和聊天式的 AI Assistant 中选择使用 GPT‑5 。

“我们用 ChatGPT 、用 GPT‑5 、也用 Codex……对 Junie 而言,常选的 LLM 之一就是 GPT‑5 。 ”—— Kris Kang

工程师们已经在把真实任务交给这些代理,并看到它们被完成。“我会把越来越复杂的任务交给以 GPT‑5 为后盾的代理,结果令我惊讶——很多任务都被成功完成了。” Kang 说。

对 JetBrains 来说,衡量标准不是单纯的速度,而是持续的工程质量。“这不仅仅是生成代码——代码必须安全、可读且易于维护。”Kang 补充道。

JetBrains 从两方面衡量影响:

  • 速度:更少样板代码、更少上下文切换、更快迭代。
  • 质量:可读、可审查、可维护的代码,而不是那些在生产中会出问题的巧技。

来自 JetBrains 的领导力经验

  • 从人们感到摩擦的环节入手:文档、测试、代码评审、交接。
  • 保护深度工作:上下文切换的成本远比打字慢要高。
  • 建立混合而非替代的工作流:由 AI 起草,人类负责设计与审查。
  • 提高基本功:清晰的意图与稳健的架构会成为倍增器。
  • 做能叠加的实验:高效迭代胜过一锤定音的证明。

“聊天会给你推进力,代理会带来质变。”—— Kris Kang

接下来会怎样

未来的工程师将更多地:

  • 设计系统
  • 指导并为代理设定护栏
  • 更高效地审查与推理
  • 用更高信心更快地交付

工作不会更少,但会更好。

“那些善于用 AI 做实验的人,会随着时间获得叠加的优势。”—— Kris Kang

加入前沿

全球已有超过 100 万家企业借助 OpenAI 取得实质性成果。联系团队:https://openai.com/contact-sales/



If you don’t write software, you may not know JetBrains.
If you do, you almost certainly use them.


The company sits behind the scenes of modern development—powering the tools used by roughly 15M professional engineers around the world (88 of the Fortune 100) and creators of Kotlin (the official programming language for Android). If you’ve opened IntelliJ, PyCharm, WebStorm, GoLand, or Rider, you’ve used JetBrains.


We sat down with Kris Kang, Head of Product at JetBrains, to explore how the team is using OpenAI models to change how developers build—not to replace what they do, but to raise the ceiling.


“Developers don’t just write code. They review it, reason about it, and design systems. AI can help with the parts beyond simply typing.”
Kris Kang, Head of Product, JetBrains



How JetBrains is adopting OpenAI



 “+15M developers use JetBrains—and now we’re bringing OpenAI into that workflow” Kang tells us that. The shift isn’t just about automation; it’s also about empowerment. It’s about protecting a dev’s flow, reducing repetitive work, and letting engineers focus on design, architecture, and judgment—the skills that give you longer leverage with AI. 


Internally, JetBrains teams are using:


  • ChatGPT
  • GPT‑5
  • Codex

Externally, JetBrains customers can choose GPT‑5 in Junie, the company’s coding agent, and in AI Assistant (for chat assistance).


“We use ChatGPT. We use GPT-5. We use Codex… one of the LLMs of choice for Junie is GPT-5.”
Kris Kang, Head of Product, JetBrains



Engineers are already delegating real tasks to agents—and seeing them completed.  “I assign increasingly difficult tasks to an agent, backed by GPT‑5—and to my surprise, many of the tasks are completed successfully” says Kang. 


JetBrains’ benchmark isn’t speed alone—it’s sustained engineering excellence. “It’s not just about generating code—it has to be safe, readable, and maintainable” Kang continues. 


JetBrains considers impact through two lenses:


Speed: Less boilerplate, fewer context switches, faster iteration.


Quality: Readable, reviewable, maintainable code—not clever output that breaks in production.


Leadership lessons from JetBrains



Start where humans feel friction: Documentation. Tests. Reviews. Hand-offs.


Protect deep work: Context switching kills more than typing speed ever will.


Build hybrid—not replacement—workflows: AI drafts. Humans design and review.


Raise the bar on fundamentals: Well-specified intention and strong architecture become a force multiplier.


Run experiments that compound: Efficient iteration beats instant proof.


“Chat gives you a lift. Agents give you a step-change.”
Kris Kang, Head of Product, JetBrains



What's next



A future where engineers:


  • Design systems
  • Guide and guardrail agents
  • Review and reason more efficiently
  • Ship faster with more confidence

Not less work—better work.


“Those who experiment well with AI will see compounding advantages over time.”
Kris Kang, Kris Kang, Head of Product, JetBrains




Join the frontier

More than 1 million businesses around the world are achieving meaningful results with OpenAI.
Contact our team




Generated by RSStT. The copyright belongs to the original author.

Source

Report Page