Cursor is a code editor — an application that software engineers use to write and edit code. It is three years old. It was founded in 2022 and until recently was largely unknown outside developer communities. In the past twelve months it has become, by some measures, the fastest-growing developer tool in the history of enterprise software.
Sixty billion dollars. Let that number sit for a moment before we try to explain it.
The headline from the week is that Cursor is in talks to raise two billion dollars in new funding at a valuation of over fifty billion dollars, forecasting more than six billion dollars in annualised revenue by the end of 2026. But it is the SpaceX number that demands attention. An option to acquire is not a valuation — it is a ceiling price that a specific buyer has decided it is worth paying. SpaceX, a company that manufactures rockets and satellites and runs a global satellite internet network, has decided that a three-year-old code editor is worth sixty billion dollars to its future.
To understand what this means, you have to understand what Cursor is actually doing — and what it is not.
Cursor is not a tool that writes code for you. It is a tool that sits alongside an engineer and acts as an intelligent collaborator: autocompleting code, explaining what a codebase does, suggesting fixes for errors, answering questions about specific functions, and increasingly — in its agent mode — taking multi-step actions across a codebase autonomously. Engineers who use it describe a qualitative shift in how they work: less time reading documentation, less time writing boilerplate, more time on the architectural decisions that require genuine judgement.
The productivity data bears this out. Cursor’s growth from roughly one billion dollars in annualised revenue to a projected six billion in twelve months is itself evidence that a lot of engineering teams have decided the tool is worth paying for. But the more interesting evidence is what SpaceX’s interest reveals about how the most technically sophisticated organisations are beginning to think about engineering capacity.
SpaceX is not buying a code editor. It is buying what the code editor represents: a significant and durable increase in the effective output of every software engineer it employs. SpaceX runs what is arguably the most complex software-hardware integration programme in the world. Falcon 9, Starship, Starlink — each involves millions of lines of code maintained by thousands of engineers across interconnected systems. An AI coding tool that makes each of those engineers meaningfully more productive is not a nice-to-have. It is a strategic asset of the first order.
The sixty billion dollar option is SpaceX saying: we believe this tool will be so embedded in our engineering operation, and so central to our competitive advantage, that we would pay sixty billion dollars to ensure we cannot lose access to it.
This is a new kind of enterprise software valuation logic. Traditional enterprise software is valued on seats, on switching costs, on the depth of integration with existing workflows. Cursor’s valuation is being built on something different: on the productivity multiplier it applies to one of the most expensive and scarce resources in the modern economy, which is skilled software engineering talent. If a coding agent makes each engineer 30% more productive — a conservative estimate, based on published studies — and if engineering capacity is a binding constraint on what a company can build and how fast it can build it, then the value of that 30% gain is not the cost of the tool. It is a fraction of the total value of what gets built faster.
This logic has implications well beyond the technology sector.
Every organisation that employs software engineers — and that is now most large organisations — faces the same calculation. Healthcare systems building patient-facing applications. Financial services firms modernising core banking. Retailers rebuilding logistics platforms. Defence contractors building the next generation of autonomous systems. For all of them, engineering capacity is a bottleneck, and a tool that expands that capacity has value that is hard to cap.
The more interesting question — and the one that Cursor’s trajectory raises without fully answering — is what happens when coding agents move beyond augmenting engineers to partially replacing the need for them. Cursor’s agent mode already handles tasks that would have required a junior engineer an hour to complete: finding a bug, understanding a codebase, making a targeted change across multiple files. Replit’s agent mode goes further, capable of building simple applications from a plain-English description with no code written by a human at any point.
We are not yet at the point where this displaces engineering roles at scale. The tasks that agents handle well are real but bounded: well-specified, well-scoped, within existing codebases. The tasks that require genuine architectural judgement — deciding how to structure a new system, choosing where the abstraction boundaries should be, thinking about failure modes that have not happened yet — remain firmly human. But the boundary is moving, and it is moving faster than most organisations are planning for.
The sixty billion dollar number is a data point, not a conclusion. One company’s option price reflects one company’s specific circumstances, its specific engineering challenges, its specific valuation of what an AI-augmented engineering team could do. It is not a universal price.
But it is a signal. It tells you that the most technically sophisticated organisations in the world are beginning to price AI’s impact on their engineering capacity not as a productivity tool but as a strategic asset — the kind of asset worth locking up at any price.
For every CIO or technology leader reading this, the question is not whether your engineers should be using tools like Cursor. They already are, or will be shortly. The question is whether your organisation is thinking about AI coding tools with the strategic seriousness that a sixty billion dollar option price implies.
Because that price is a market’s answer to the question: how much is it worth to make your engineers significantly more effective? SpaceX has given its answer. Most organisations have not yet given theirs.
