AI is changing weather prediction in two ways at once. It is making forecasts easier to generate, because newer deep learning techniques can run simulations on laptops instead of supercomputers. It may also make weather data easier to use inside real business decisions.
WindBorne Systems is trying to build a company at the center of that shift. The startup, founded in 2019, collects atmospheric data with long-flying weather balloons, feeds that information into a forecasting model, and is now preparing for a broader commercial push after raising a $37 million Series B round.
A funding round built around better forecasts
WindBorne Systems raised a $37 million Series B round, CEO John Dean told TechCrunch. The round was co-led by Khosla Ventures and Galvanize, with additional investments from Translink Capital, Lux Capital, and previous investors.
The financing values the company after this round of funding at $250 million. For WindBorne, the investment is not only about building a larger sensing network. It is also about proving that better forecasts can become a useful product for organizations that need weather intelligence but have not always had an easy way to act on it.
The company began with a plan to gather a new kind of weather dataset using low-cost weather sensors and endurance balloons. That original idea has become more valuable as AI weather forecasting models have improved over the last four years.
Before those advances, making private forecasts at this level was much harder. The source article explains that most private companies previously could not do it because simulating the atmosphere required expensive supercomputers. AI has lowered that barrier, giving companies like WindBorne a path to make forecasts from their own data.
How WindBorne collects weather data
WindBorne currently has 20 launch sites around the world and about 600 balloons in the air at any given time. Those balloons collect data in places that are difficult to reach, including the eye of a typhoon.
The company is also beginning to deploy aerial sensor packages that can fall into the ocean and keep collecting measurements as floating buoys. That expands the company’s data-gathering approach beyond the air and into ocean-based observations.
Dean calls the company’s proprietary dataset a “planetary nervous system.” That dataset is central to WindBorne’s strategy because it creates a moat around its weather model. The model also uses datasets produced by government weather agencies around the world.
Dean said the balloon data has already shown value inside forecasts. “We demonstrated that when you add balloons to the forecast, you get more accurate forecasts, and the value per data point is much stronger than satellites,” Dean said. “We’ve also been growing revenue while we’re doing that, so that de-risked the demand signal to VCs.”
That point matters for both the technology and the business. If WindBorne’s balloon data improves forecast quality, the company has something more specific to sell than a generic weather model. It can argue that its sensing network gives customers information they would not otherwise have.
Government customers came first
WindBorne’s main customers today are government agencies. The U.S. National Weather Service buys the company’s data. The U.S. Air Force and U.S. Navy are also paying WindBorne through research partnerships.
One of those efforts involves forecasting models that can run onboard ships when connections to the rest of the world may be intermittent. That use case shows why local forecasting capability can matter: a forecast is more useful when it can still function in environments where connectivity is unreliable.
Government demand is a natural fit for a company built around weather sensing. Public agencies already understand how to use weather data, and they have established workflows for putting it into forecasts and operations.
That same familiarity is harder to find in the private sector. Many sensing companies have discovered that collecting valuable data is only part of the challenge. Customers also need the experience, tools, and internal processes required to turn data into decisions.
The harder move into commercial markets
WindBorne’s next step is commercial business. For now, that effort is mainly focused on investment funds that use weather data to predict commodity prices and other business outcomes.
The new funding will support several priorities. WindBorne plans to spend on compute, work on replacing the balloon network’s satellite communications with a mesh radio network, and build out its go-to-market team to expand its private-sector customer base.
That expansion will not be automatic. Over the last decade, several startups have tried to scale sensing businesses such as earth observing satellite networks. Many found it difficult to break through with private customers because the value of the data depended on specialized knowledge and established workflows.
Private weather forecast companies already exist, but the source article notes that they mostly make money in specific areas:
- Repackaging or refining government forecasts for news media
- Serving specialized needs such as plane de-icing and ship routing
- Providing data for speculators
AI could change the size and shape of that market. If better models make forecasts more accurate, and AI tools make those forecasts easier to connect to business decisions, more companies may have a reason to buy weather intelligence directly.
Saloni Multani, a partner at Galvanize who co-led the round, framed the opportunity around that shift. She said the private weather market has been limited because “integrating weather forecasts into broader business decision-making has traditionally been expensive and difficult. We think AI changes that equation. Better forecasts make the effort worthwhile, and AI makes it much easier to connect those forecasts to the decisions businesses are trying to make.”
That is the core test for WindBorne. The company has funding, a global balloon network, government customers, and an AI forecasting model. Its next challenge is turning better weather prediction into a product that private organizations can use without needing to become weather experts themselves.