Kristel Cocoli

cyber systems engineer | technical creator

New on X, bear with me.

Instagram: @kristel_tech Tiktok: @kristel_tech

[Kristel Cocoli @kristel_tech]

@MaiVangSWE PLEASE DO

[Kristel Cocoli @kristel_tech]

Developed my own J.A.R.V.I.S in JetBrains Air.

Used Codex (Astra) + Junie Lite + handwritten code where I saw fit.

WYT? What other features should I add?

[Kristel Cocoli @kristel_tech]

Hi! You can actually use a Claude sub on Air, in the Terminal UI :D Air opens it as a full-screen editor tab, & you get the real Claude Code experience inside the IDE, with your existing plan and no extra keys. (For Claude Agent in the chat UI on the other hand, you will need to have a JetBrains AI subscription or use your API keys)

[Kirill Skrygan @kskrygan]

We connected IntelliJ IDEA’s code intelligence to Claude Code and Codex, then tested the setup with Netflix.

The first public results are out.

Typical efficiency gains: 20–30%.
Best code-navigation case: ~65%.

These results come from two real codebases. The efficiency metric averages token savings and time savings.

Correctness stayed stable or improved.

Here’s what we actually built.

The code engine behind IntelliJ IDEA ran headlessly, giving agents access to its capabilities through a CLI backed by a daemon.

LSP alone wasn’t sufficient.

Agents needed a convenient way to invoke the tools. The backend also needed to stay running and keep the project index warm between calls.

The idea is straightforward.

An IDE has already mapped the structure of your code and its dependencies — including what’s inside compiled libraries.

An agent can use that knowledge instead of spending time and tokens reconstructing it.

But the most interesting finding was which models benefited most.

The strongest models saw the biggest gains.

Especially at higher reasoning effort.

In these experiments, more capable models were also better at taking advantage of code intelligence.

My takeaway: better models and better tools aren’t competing bets.

You don’t have to choose between investing in model intelligence and giving agents better access to your codebase.

The two can reinforce each other.

Huge thanks to @pbakker and the Netflix team for working with us on this!

[Jakub Chrzanowski 🥑 @hszanowski]

A user says your plugin froze their IDE. You can't reproduce it.

JetBrains Marketplace now shows freeze reports linked to your plugin, with excerpts and full thread dumps.

I made a short walkthrough of the setup and investigation:
https://t.co/dOPI5v8oLU

[Kristel Cocoli @kristel_tech]

@jcraane @jetbrains I had 3 hours to prepare the project so I did not play around enough, but now that I have more time I will stress test them more!
For Opus: effort high
For Gemini: thinkingLevel: high

[Kristel Cocoli @kristel_tech]

@Matt27659906 @jetbrains High. For both.
For Gemini modify it in its settings.json

[JetBrains @jetbrains]

Gemini 4 Argon is officially live!

@kristel_tech tasked two agents with building a web-swinging delivery game in a hyper-realistic night city:

• Gemini 4 Argon: $54.41 | 23 mins
• Opus 5.5: $56.97 | 2h 20 mins

Compare their performance yourself in JetBrains Air or Junie.

[Kristel Cocoli @kristel_tech]

@kskrygan THIS IS EXCITING 🔥🔥🔥

[Kirill Skrygan @kskrygan]

Holy moly, we’ve seen a lot of interest in JetBrains Air since the announcement!

-Air in IDEs: for developers
-Air Teams: for team leads & engineering managers
-Air Governance: for CTOs & engineering directors

Here’s how to try these things:
Links below 🧵

[Kirill Skrygan @kskrygan]

With Air in our IDEs, you can easily work with Claude, Codex, and 50+ other agents.

But we also bundled Junie Lite right into Air — so you always have a baseline agent ready to go.

And for something that’s free, it’s surprisingly good.

Check it out, share your feedback.

[JetBrains @jetbrains]

Running multiple agents in parallel? Want the freedom to choose your agent?

Try JetBrains Air in your IDE – now in Early Access.
Get started for free*, with Junie Lite configuration.

*𝘛𝘦𝘳𝘮𝘴 𝘢𝘯𝘥 𝘤𝘰𝘯𝘥𝘪𝘵𝘪𝘰𝘯𝘴 𝘢𝘱𝘱𝘭𝘺

[JetBrains @jetbrains]

The JetBrains Air in IDEs is here. Run your agents of choice in parallel and review their changes with JetBrains’ code intelligence, all in the IDE you already use.

Check below why this is the next step for JetBrains Air. 👇

[Kirill Skrygan @kskrygan]

We've reimagined JetBrains IDEs to become the ultimate agentic workbench.

Air is an open hub for agents inside your IDE:

-Any agents to run (Claude, Codex & many more)
-Agents are more efficient with IDE intelligence
-Free Junie Lite agent bundled
-Cloud runs via Air Teams
⬇️

[Kristel Cocoli @kristel_tech]

@s_sophia68 @jetbrains @OpenAIDevs If u connect it to Blender MCP, u can do so much more, partially with AI, and partially manually (it is fun, I ve tried it. Takes more time ofc but worth it)

[Kristel Cocoli @kristel_tech]

@lifemademe @jetbrains @OpenAIDevs In this particular one the difference was pretty obvious. But I should run more tests 😅

[Kristel Cocoli @kristel_tech]

@TatataToddC @jetbrains @OpenAIDevs Price-per-quality seems really good!
I need to check code quality now (can t speak with confidence yet). In terms of visuals it s pretty pretty good for the price!

[OpenAI Developers @OpenAIDevs]

GPT-6.1 Sol is here.

Upgraded with stronger agentic coding and computer use, near-Astra performance, and cached input at a 95% discount to standard input pricing.

GPT-6.1 Sol is built for complex refactors, deep codebase investigations, and long-running agents across apps.

[Kristel Cocoli @kristel_tech]

@victor_vibing @jetbrains @OpenAIDevs I gotta stress test 6.1 more to form an opinion. But what about cost-wise? Have you compared yet? Because I m impressed what can be produced with Sol 6.1 for the cost

[Kristel Cocoli @kristel_tech]

@BharadwajaSimh1 @jetbrains @OpenAIDevs Inception-level posts