Netflix plugged JetBrains& 8217; headless code intelligence engine – the same one behind IntelliJ IDEA – into Claude Code and Codex, then benchmarked it against a variety of scenarios. The details and numbers are available below. At a glance Company Netflix – Application Development Platform (ADP), GenAI Platform,…
Trained with reinforcement learning in real environments, Mellum2.1 is built for coding agents and fast sub agents that run on your own hardware. Today, we& 8217;re releasing Mellum2.1, the next version of the 12B mixture of experts model we open sourced in June. The architecture hasn& 8217;t changed since version 2:…
Last week, we shared our point of view on where we think agentic development is headed and introduced JetBrains& 8217; answer to it: JetBrains Air, an open system of products for agentic development. Today, we& 8217;re launching the Early Access Program for Air in JetBrains IDEs. It& 8217;s available now as a plugin…
Part 1: Parsing, chunking, and vectorization Some time ago, we set out to build the best semantic code search platform we could: a RAG pipeline that gives LLM agents precise, citable evidence from real repositories instead of whatever grep happens to surface. The eventual solution was Air Context. We got it working,…
Air gets a new multiproject view: open several projects in one window and run agents across them in parallel. The sidebar groups tasks by project, so you always see which agent is working on which project – no more window juggling for multirepo work. Markdown files now render as formatted text while you edit – [&…
Over the past six months at JetBrains, our AI development expenses have increased roughly 10x. When the costs started rising, of course we noticed – and realized that we simply didn& 8217;t know how to control them systematically. We didn’t know because our developers don’t just use the AI tools we build. They decide…
Part 3 of a series where we take public & 8220;token saver& 8221; add ons for coding agents and run the same paired A/B benchmark against each of them. Part 1 was the caveman skill (advertised −65%, measured −8.5%). Part 2 was rtk (advertised −60–90%, measured +7.6%). We ran 80 paired tasks to test the ponytail skill…
The new release of JetBrains Air brings support for GitHub Copilot, OpenCode, Pi, Cline, and other ACP compatible agents. It also adds IntelliJ powered navigation and diagnostics for Java and Kotlin and runs Windows tasks in Docker containers. Try Air for Mac, Linux and Windows Bring your own agents and harnesses Air…
Today, we’re launching JetBrains Context, a new repository intelligence layer that helps coding agents work more efficiently and produce higher quality results on complex codebases. As part of the JetBrains AI for Teams and Organizations rollout, JetBrains Context is now available in early access at no additional…
Does & 8220;rtk& 8221; reduce Claude Code token usage? Part 2 of a series where we take public “token saving” add ons for coding agents and run the same paired A/B benchmark against each of them. Part 1 was the caveman skill (advertised −65%, measured −8.5%). TL;DR: rtk advertised saving: 60–90%. Measured on real…
Agentic coding benchmarks are getting closer to real world software development. For Kotlin teams, the most important question is how reliably AI agents can complete end to end Kotlin tasks, from reading an issue to producing a solution that passes validation. We’re taking the first step in addressing that gap by…
We’re about to start rolling out a new set of AI capabilities that provide shared context, reusable agentic workflows, organization level governance, and cost control for software production. Developers use different AI tools depending on the task – from JetBrains IDEs to terminal based agents such as Claude Code,…
