Wallace Hogsett has spent more than two decades working in weather, from studying hurricane modeling and developing software at the National Hurricane Center to working with insurance companies on weather risk. Across those roles, he has encountered the same problem from a number of angles: There is plenty of weather data, but turning it into a decision can be difficult.
Where hurricanes are concerned, that wealth of data has even helped bring about global forecasting models that have taken decades to develop. But while many public weather services—NOAA, for example—provide information for broad audiences, specific, actionable insights relevant to individual organizations can be harder to come by.
With his new startup, HurriLab, Hogsett aims to provide the analysis layer for those organizations confronting more specific hurricane risks; the company will help them sort through all that available data, pull from the most useful sources, and turn it into actionable information for individual decisions.
“Me starting this company is kind of like leveraging all of these things I’ve heard over the years, and saying [that] I think we can do things a little better to give people information that helps them actually make the decision they need to make,” Hogsett said.
Problem and solution
To illustrate the headache decision-makers can face when a hurricane threatens one of their assets, Hogsett recalled a commodities trader he spoke with in Chicago. The trader is actually a meteorologist himself, but not a hurricane expert. So, Hogsett asked him to talk through how he might decide what trades to make when a storm could affect the commodities he follows.
“[The trader] said, ‘Well, I’ve got 25 tabs open, and I’m refreshing them all,’” Hogsett said. “Looking at, you know, an observation station, a satellite, the latest run of the Google model or the NOAA model.”
Even as a meteorologist making high-stakes decisions, the trader was manually moving between roughly 25 sources to piece together an idea of what the storm might do. This time-intensive workflow illustrates the problem HurriLab is trying to solve: the The data exists, but it’s rarely concentrated in a way that usefully informs specific decisions.
The solution begins with HurriLab pulling in dozens of weather datasets—including hundreds of forecasts from ensemble systems around the world, alongside real-time observations from satellites. Those sources can have different strengths: Some models may better capture a hurricane’s track, while others may provide useful information about the same storm’s intensity. Satellite data, meanwhile, can provide information about where a storm currently is, its size and what winds look like at the surface.
“With the infrastructure layer, as I call it, I’m pulling in dozens of different data sets, homogenizing them, taking what’s good and discarding what’s bad, and ultimately producing this unified, global, probabilistic data set,” Hogsett said.
Notably, HurriLab’s product is not meant to provide a forecast as the weatherperson would on television. Rather, the startup is in the business of providing probabilities for specific outcomes.
For example, rather than informing an oil refinery that the hurricane track is on course to hit one of its facilities, HurriLab could provide the probability that winds at that location will exceed a certain threshold: something along the lines of, “There’s a 60% chance that winds will reach 50+ mph in the next 24 hours.” The refinery could then use that probability alongside its own operating procedures and costs to decide when to shut down a facility, evacuate workers or take other precautions.
Right now, Hogsett demonstrates the data through a map-based dashboard. But the dashboard isn’t necessarily the product itself. Hogsett said the underlying data could eventually be delivered through an API or used to power other applications.
Users and rollout
Hogsett sees HurriLab’s initial customers as organizations large enough that hurricanes can cause them million-dollar problems, but which do not have their own teams of meteorologists.
“There are companies that have entire emergency operations centers, but they don’t have meteorologists,” Hogsett said. “And so they end up with the 25 tabs open.”
A port, for example, may need to know when winds will become too strong to safely operate cranes, or which direction winds will come from to determine how to moor boats. A large retailer might need to know when highways will become impassable or where to position inventory before a storm.
QUICK BITS
Startup: HurriLab
Founder: Wallace Hogsett
Founded: 2026
Team size: 1
Location: Wilmington
Website: hurrilab.com
Funding: Bootstrapped
As of this writing, HurriLab is still at an early stage. Hogsett incorporated the company earlier this year and is spending much of his time talking with prospective customers while continuing to build the underlying infrastructure.
By next hurricane season, his goal is to be working closely with at least five customers through paid proof-of-concept projects—both to prove HurriLab can provide value beyond freely available weather data and to better understand how customers want to use it.
Over the long term, Hogsett sees HurriLab as part of a broader shift happening in meteorology as AI and other technologies change how weather information is produced and used.
“The real why behind all this is I think the weather world is ready for some pretty big changes,” Hogsett said.
Hogsett recently represented HurriLab while speaking to a room full of startup ecosystem leaders and blue innovation pioneers about commercializing the blue economy at the 2026 Ocean Innovation Conference in Wilmington.
