The story behind Cortex, a privacy-first AI IDE
Cortex is a privacy-first cross-platform AI IDE from Pasons Group. Prompts and code go to your provider, never through our servers. Free with your own keys.
The short answer: Cortex is a native AI IDE for Windows, Linux, and macOS from Pasons Group (Dubai, established 2003). Your prompts and code go from your machine to the model provider you choose. They do not pass through Cortex servers. The IDE is free with your own API keys. This page is for developers who want an agentic editor without routing a private repo through someone else's proxy.
What is Cortex, and who is it for?
Cortex is a privacy-first agentic IDE for Windows, Linux, and macOS. Use it when you write production code on your own machine, you hold your own API keys, and you do not want chat or source files proxied through the IDE vendor. It is not a cloud workspace and it is not an Electron wrapper that tells you to install a container.
Cortex shipped its first public release, v2.7.0, on July 9, 2026. It is built and backed by Pasons Group, a diversified business group established in 2003 and headquartered in Dubai, UAE. This is the story of why it exists, who stands behind it, and the design rule at its core: your prompts and your code should never touch anyone else's server.
The company behind Cortex
Pasons Group has grown from a single supermarket and catering service in 2003 into a group of companies spanning hypermarkets and supermarkets, shopping malls, restaurants and catering, wholesale and mobile trading, e-commerce, and real estate: more than 70 outlets serving millions of customers across the UAE. Cortex AI IDE is the group's technology venture: the same long-term, customer-first philosophy, "Your Satisfaction Is Our Priority", applied to developer tools.
- Established
- 2003
- Headquarters
- Dubai, UAE
That backing matters for a practical reason: an IDE is infrastructure. Developers who build their daily workflow around a tool deserve to know there's an established organization behind it, one with decades of operating history and a name it protects, not an anonymous app that might vanish next quarter. Cortex is free to use with your own API keys, and it's run with the patience of a group that has been serving customers since 2003.
Why build another AI IDE at all?
Because the tools we wanted to use kept failing the same tests. They treated the OS as an afterthought: Electron wrappers and "just use a container." They asked developers to trust a subscription with unbounded inference costs. And above all, they asked everyone to route code and prompts through their servers and take their word for what happened there. The 2026 consolidation wave made the trust question sharper, not softer: the major AI IDEs are now business units inside giants, and your relationship is with whoever owns them this quarter.
Cortex takes the opposite path: the capabilities the big tools have (agentic loops, multi-provider models, MCP, semantic search) built into one native IDE for Windows, Linux, and macOS that belongs to its users. Free to use with your own keys. No subscription required to code.
How does Cortex keep prompts off its servers?
The architecture is the privacy policy. Cortex is the wrapper; the model is the brain. When you chat, your prompt and code go directly from your machine to the LLM provider you chose (Anthropic, DeepSeek, OpenAI, MiMo, Qwen, OpenRouter) over TLS. There is no Cortex proxy in the middle. Architecturally, our servers cannot read your prompts or code, which is a much stronger guarantee than "we don't."
- Downloading is anonymous. Grab the installer and go, no account, no sign-up wall.
- Signing in reveals exactly one thing: your email. That plus your plan is all the server knows. What you're building, what you're asking, which files you open, nothing is tracked, because the tracking was never built.
- Your API keys never leave your machine: stored in your operating system's secure credential store, encrypted at rest, with an AES-256-GCM local backup.
| Decision | Typical cloud AI IDE | Cortex |
|---|---|---|
| Where prompts go | Vendor proxy, then the model | Your PC to the provider you pick |
| API keys | Often stored or billed by the vendor | Your OS secure credential store |
| Native install | Electron or container-first | Native Windows, Linux, and macOS, no container |
| Must you subscribe? | Usually, to get the agent | No. BYOK is enough to code |
| Who this is for | Teams already inside that vendor | Developers who keep keys local |
Why is "trust us" not good enough?
Because you cannot audit a remote proxy you do not run. The industry has already seen what happens when developers take a vendor's word for it: popular AI coding tools have been caught sending far more of a project than any coding task required, with the data controls users relied on making no difference. The lesson is general: when your tool routes data through a vendor, you're trusting configuration you can't see, changeable remotely, by people you'll never meet.
The only architecture that survives that lesson is the one where the data never leaves in the first place. That's Cortex's design, and it's why the code is being opened for inspection: core components are public on GitHub today, with full open-sourcing planned as each part clears security review, so third parties can verify the no-tracking claim instead of taking ours on faith. Fitting, for a product whose whole thesis is verify, don't trust.
What did we learn shipping Cortex in public?
Development ends the day real users arrive; then the real education starts. These notes are from our own test machines and from the public changelog, not from a marketing brief. The hero above is this article's own image; we do not reuse the homepage product shot here.
- July 9: v2.7.0. Launch: agentic editing, BYOK for five providers, full editor, terminal, project memory.
- July 10-11: v2.7.1 through v2.7.6. Reality arrives: saved API keys silently ignored after install, panels rendering blank on certain GPUs, a ghost "capsule" window. Five releases in three days, fixing what testing on one machine can't catch, other people's machines.
- July 12: v2.8.0. MCP support, Live Preview, and performance work that took a 40-second chat-open freeze to near-instant.
- July 13: v2.8.1. Core source code lands on GitHub, and the loop engine's token/dollar budgets become real enforcement, not a checkbox.
- July 14-18: v2.8.2 through v2.8.6. Antivirus stutter fixed with background-thread writes, vision-model image handling rebuilt, Claude Sonnet 5 and Kimi K3 added within days of their releases.
The hardware and antivirus matrix is the boss fight. Features are easy compared to running correctly on every GPU driver and under every real-time scanner. That is original experience, not a slogan: we measured the 40-second freeze, shipped the fix the same week, and left the note in the changelog.
Should I subscribe, or is BYOK enough?
BYOK is enough to code. The IDE is free with your own API keys, forever. You pay model providers directly, and at BYOK prices that's dramatically cheaper than fixed AI-IDE fees. The $10/month Pro plan exists for the things that genuinely need servers, hosted web search, semantic embeddings, OCR, MCP connections, and as a way to support the project if you believe in it. Subscribe when you need it, cancel when you don't, and the IDE keeps working either way. No pressure, no forced auto-renewal games. Pasons Group built its name on putting the customer first; Cortex's business model has to live up to that name.
What's next
The roadmap runs on the same loop as the product: real user reports in, verified fixes out, every release note answers an actual email. If you want to pressure-test it, download Cortex, hand it a real task, and tell us what breaks at [email protected]. Anonymous downloads mean we won't know you're there, so the emails are how we learn.
Frequently asked questions
What is Cortex AI IDE?
Cortex is a native AI coding IDE for Windows, macOS, and Linux. An agent can read, edit, run, and verify your code. You bring your own API key, and prompts go from your PC to the provider you choose, not through a Cortex proxy.
Who is behind Cortex?
Pasons Group, a Dubai-headquartered business group established in 2003. Cortex is the group's developer-tools venture, not an anonymous app with no operating history.
Do my prompts go through Cortex servers?
No. Chat and code go directly from your machine to the LLM provider you selected, over TLS. Cortex servers cannot read prompts or source because they are not on that path.
Should I subscribe to use Cortex?
No. The IDE is free with your own API keys. The optional Pro plan is only for hosted extras such as web search, embeddings, OCR, and MCP connections.