Cursor is rewriting how software gets built. The AI-native code editor, developed by Anysphere, combines VS Code's familiar interface with proprietary models that generate, edit, and debug code across entire projects. Serving the majority of the Fortune 500 and over 50,000 teams globally, Cursor has become the fastest-growing developer tool in a generation.
Founded 2022 | HQ: San Francisco | Funding: $3.3 billion | Valuation: $29.3 billion
MM Verified
Overview
Cursor was built by Anysphere, a company founded in 2022 by Michael Truell, Sualeh Asif, Arvid Lunnemark, and Aman Sanger, four MIT graduates who turned down Big Tech offers to build an AI-first code editor. The team graduated from OpenAI's accelerator programme in 2023 and launched the first version of Cursor in March of that year.
The growth trajectory has been extraordinary. Cursor crossed $1 billion in annualised recurring revenue by late 2025, making it one of the fastest enterprise SaaS products to reach that milestone. The company closed a $2.3 billion Series D in November 2025 at a $29.3 billion valuation, led by Accel and Coatue with strategic participation from Nvidia and Google. That round came just five months after a $900 million Series C that valued the company at $9.9 billion.
The product is built as a fork of Microsoft's open-source VS Code, extended with proprietary AI capabilities including Composer, Cursor's custom mixture-of-experts model that runs four times faster than comparable LLMs. The editor supports multi-file editing, codebase-wide context awareness, and background agents that can execute tasks autonomously.
What We Like
Composer model is a genuine technical differentiator. While most AI coding tools rely on third-party models, Cursor built its own. Composer 1.5 uses reinforcement learning with adaptive thinking that adjusts reasoning depth based on task difficulty. The result is a model purpose-built for code generation that runs faster than general-purpose alternatives, with 20x RL scaling improvements over the previous version.
Codebase-wide context that actually works. Cursor indexes your entire project and uses that context when generating or editing code. You can reference files with @-mentions, ask questions about your codebase in natural language, and make changes across multiple files in a single operation. This is materially different from autocomplete tools that only see the current file.
Background agents for asynchronous development. Pro+ and Ultra tiers unlock background agents that run multi-step tasks independently: executing test suites, refactoring across files, or implementing features while you work on something else. This moves Cursor from "AI assistant" to "AI teammate" territory.
Pricing that respects individual developers. A generous free tier (Hobby), a $20/month Pro plan, and clear upgrade paths to Pro+ ($60) and Ultra ($200) make Cursor accessible to solo developers and small teams. The pricing transparency is unusually strong for a product at this stage of growth.
What to Watch
VS Code dependency is a double-edged sword. Cursor's VS Code foundation means extensions, keybindings, and settings port seamlessly. But it also means Cursor is downstream of Microsoft's editor roadmap. Any breaking change in VS Code propagates to Cursor, and developers who rely on proprietary VS Code features (like GitHub Copilot's native integration) may face friction.
Usage limits can frustrate power users. Pro plan users get 500 fast agent requests per month. Heavy users report hitting limits mid-sprint, forcing a switch to slower models or an upgrade. The limits have been increasing, but the model remains consumption-based, not unlimited.
Enterprise tier is custom-priced. While individual and team pricing is transparent, Enterprise pricing requires a sales conversation. Large organisations evaluating Cursor against GitHub Copilot Enterprise will need to negotiate terms, which slows procurement.
Pricing and Deployment
Cursor offers five tiers: Hobby (free, limited completions), Pro ($20/month, 500 fast agent requests), Pro+ ($60/month, 3x capacity plus background agents), Ultra ($200/month, high-volume), and Teams ($40/user/month). Enterprise pricing is custom. Deployment is a desktop application for macOS, Windows, and Linux. Settings and extensions sync from VS Code with one click.
Compliance and Security
Cursor holds SOC 2 Type II certification with annual third-party penetration testing. The platform is GDPR and CCPA compliant. Enterprise customers can enforce Privacy Mode organisation-wide, which activates a zero data retention agreement with upstream model providers. All data is encrypted with AES-256 at rest and TLS 1.2+ in transit. Privacy Mode compliance is verified every five minutes at the organisational level.
Verdict
Cursor is the right choice for development teams that want AI deeply integrated into their editing workflow, not bolted on as a chat sidebar. The proprietary Composer model, codebase-wide context, and background agents represent genuine technical advantages over autocomplete-style alternatives. Solo developers and small teams get exceptional value at the Pro tier. Organisations in highly regulated industries should evaluate whether Privacy Mode meets their compliance requirements before committing, and large enterprises should expect a custom pricing conversation. Cursor's trajectory, from zero to $1 billion ARR in under three years, suggests the product will continue evolving faster than almost anything else in the developer tools market.
Try Cursor: cursor.com
How we scored it
| Criterion | Score | Notes |
|---|---|---|
Compliance & Security 15% weight | 4.0 | SOC 2 Type II, GDPR, CCPA; Privacy Mode with zero retention |
Documentation 15% weight | 4.0 | Clean docs site; growing but not yet comprehensive for enterprise |
Ease of Setup 10% weight | 4.5 | VS Code fork with one-click migration of extensions and settings rated 4.5/5 for Integration Ease. |
Integration Flexibility 10% weight | 4.5 | VS Code fork; one-click migration of extensions and settings |
Support Quality 10% weight | 3.5 | Community-first support model; enterprise support still maturing |
Scalability 10% weight | 3.5 | Pro plan caps at 500 fast agent requests/month; power users report hitting limits mid-sprint, forcing upgrades. |
Pricing Transparency 10% weight | 4.0 | Published tiers for individuals and teams; enterprise requires sales |
Pros
- Composer model is a genuine technical differentiator
- Codebase-wide context that actually works
- Background agents for asynchronous development
- Pricing that respects individual developers
Cons
- VS Code dependency is a double-edged sword
- Usage limits can frustrate power users
- Enterprise tier is custom-priced
Sources
Editorial disclaimer: Reviews reflect the independent editorial assessment of Major Matters and are not sponsored or endorsed by the companies reviewed. We recommend conducting your own evaluation to determine whether any product is the right fit for your specific requirements.
Charlie Major is a Product Development Manager at Mastercard. The views and opinions expressed in Major Matters are his own and do not represent those of Mastercard.