Podcast Summary: Inevitable – Improving Weather Forecasting with WindBorne
Podcast: Inevitable (an MCJ podcast)
Host: Cody Simms
Guest: John Dean, Co-Founder & CEO of WindBorne Systems
Date: April 7, 2026
Episode Overview
This episode delves into the overlooked yet critical infrastructure behind weather forecasting—specifically, the role of weather balloons and how WindBorne Systems is revolutionizing data collection for forecasting. Cody Simms and John Dean discuss the shortcomings of legacy weather balloon technology, the advances WindBorne has made with long-duration, AI-powered systems, and the implications for everything from hurricane prediction to grid management. The conversation explores the intersection of physical and digital innovation and examines why the advancement of real-world data gathering is as vital as improvements in AI models.
Key Discussion Points & Insights
1. The Invisible Backbone of Weather Forecasting
[02:22–04:36]
- Most people don't consider where weather data comes from. John Dean explains that nearly all 7–8 billion people on Earth rely on weather forecasts, but few think about the infrastructure behind them.
- Traditional weather observation: Weather forecasting relies on three main observation types:
- Surface stations
- In situ atmospheric observations (primarily weather balloons)
- Satellites
- Limitations of traditional balloons: Standard latex “radiosonde” weather balloons float for two hours, rise, pop, and return—providing only a one-off vertical atmospheric profile.
“A two hour full life... as they rise, the air pressure decreases, balloon gets bigger and bigger, eventually it bursts and comes back down and you get one...sounding of data. It's a vertical profile of the atmosphere.” — John Dean, [04:39]
- Satellites: Have enabled massive gains but can't provide high-accuracy ground truth, especially below clouds or just above the ocean.
2. Why Forecasts Struggle (and How Better Data + AI Help)
[06:25–07:31]
- AI is helping, but data is still the bottleneck:
“You can have the smartest AI model in the world, but if it doesn’t know the initial conditions accurately, you’re going to have inherent error.” — John Dean, [06:44]
- Intrinsic unpredictability: Even with perfect AI and data, weather’s chaotic nature means some error—and public complaints—will always remain.
3. Government and Funding Landscape
[10:09–12:34]
- Public-private partnerships are growing: Government agencies have historically led weather services, but companies like WindBorne are expanding private roles.
- Cuts to NOAA: Recent US government cuts have hit academic and climate research, not the core forecasting infrastructure.
“The thing I’m actually most worried about is ... 10 years in the future when you have a missing generation of scientists and engineers.” — John Dean, [11:26]
4. WindBorne’s Origin Story & Why Their Balloons Matter
[12:34–16:25]
- Idea genesis: WindBorne started as a fun student project—making weather balloons last longer.
“It takes a lot of smarts to design something dumb that works. And that is the philosophy of Windborne.” — John Dean, [14:41]
- Technological inflection points:
- Recent improvements in batteries, sensors, and satellite comms now allow lightweight, long-duration, low-cost balloons.
- Example: The COVID-era need for shipping cold vaccines drove innovations in low-temperature, rechargeable batteries, allowing balloons to stay aloft for months.
5. What WindBorne Balloons Do Differently
[16:25–19:33]
- Long-duration flights: Balloons can now fly for weeks or even months (average two weeks), covering vast and remote areas.
- Global “constellation” (Atlas): Hundreds of balloons (250+ at a time as of the episode) are navigated globally from multiple permanent sites.
- Balloons collect core atmospheric data—temperature, pressure, humidity, wind. The advantage: Unprecedented global coverage, including oceans, the Arctic, and hurricane interiors.
“Before windborne, we only had adequate weather observations over 15% of the planet. That’s where there’s land and enough humans to launch balloons every day.” — John Dean, [19:43]
6. Safety and Airspace Incidents
[20:17–24:20]
- Recent airspace incident: A WindBorne balloon was involved in a commercial airplane windshield strike—no depressurization or catastrophic damage, similar to bird strikes.
- Regulatory boundaries: Balloons below six pounds are not considered a major risk; WindBorne is proactively pushing for higher safety standards and transparency.
“I’m a tent person. I have a deep attachment to the truth. ... I would rather not do what Windborne’s doing than make a significant safety hazard.” — John Dean, [23:25]
7. The Business Model
[24:34–27:53]
- Three revenue layers:
- Data as a service: Selling subscription data feeds from the constellation, mainly to government agencies.
- AI-powered weather forecasting: Building models with WindBorne and public data—offering tailored forecasts and APIs.
- Insights products: APIs and tools for commercial clients (e.g., hedge funds, utilities, grid operators).
8. Paradigm Shift in Global Weather Intelligence
[27:53–30:14]
- Comprehensive picture: WindBorne isn’t introducing wholly new types of data, but vastly expands the spatial and temporal coverage, especially over previously unmeasured regions and critical areas for severe weather genesis.
- Utility grid operations: Better forecasts can halve outage restoration times by enabling pre-deployment of repair crews.
9. Technological Tailwinds and Future Ambitions
[31:49–34:48]
- Physical and digital innovation: Advances in comms chips, mesh networking, and custom AI-driven chip design will drop balloon costs from $1,000 each to ~$20, enabling planetary-scale coverage.
- Vision: Move from 250+ to potentially 1,000,000 balloons in a constantly updating “planetary nervous system.”
“If there is a future strike to [a] Windborne balloon, it'll be ... like scraping a bug off your windshield. And the cost per balloon will go from $1,000 to $20.” — John Dean, [33:43]
10. Call to Action & Mission
[34:48–37:01]
- Talent recruitment: WindBorne is hiring across the technical spectrum.
- Mission clarity: Ground-truth physical data is resistant to digital disruption and essential for any future modeled by AI.
“Improving weather forecasts by using AI is just a pure net good for the world.” — John Dean, [35:40]
Notable Quotes & Memorable Moments
- “It takes a lot of smarts to design something dumb that works. And that is the philosophy of Windborne.” — John Dean, [14:41]
- “Improving weather forecasts by using AI is just a pure net good for the world. There is no downside in replacing an expensive to run physics based model with a lower cost, lower power usage model.” — John Dean, [35:40]
- “I would rather not do what Windborne's doing than make a significant safety hazard.” — John Dean, [23:25]
- Describing the future cost drop: “It'll be like, you know, scraping a bug off your windshield.” — John Dean, [33:43]
Important Timestamps
- [02:22–04:36] — The invisible infrastructure behind global weather forecasting
- [06:44] — Why smarter AI alone isn’t enough without better data
- [10:09–12:34] — Government funding landscape and public-private partnerships
- [12:34–14:41] — WindBorne's origin and philosophy ("smart, dumb design")
- [17:30–19:33] — Coverage over oceans, hurricanes, and atmospheric rivers
- [20:17–24:20] — Airspace safety, the recent plane incident, regulatory standards
- [24:34–27:53] — WindBorne's layered business model
- [31:49–34:48] — The impact of ongoing advances in chip, battery, and comms tech
Tone & Speaker Dynamics
The discussion is accessible and lightly humorous, with John’s engineer’s pragmatism and humility shining through. Cody prompts with curiosity and seamlessly translates technical points to broader implications. The tone is optimistic about technology’s power for public good and persistent in advocating for safety and transparency.
Conclusion
WindBorne Systems is transforming weather observation by harnessing recent advances in batteries, sensors, and AI to vastly expand global, in situ atmospheric data—especially over oceans and the upper troposphere where it’s been historically lacking. Their data-rich, AI-enhanced forecasts are already producing meaningful improvements (like halving hurricane track error) and aim to support everything from disaster response to grid management. Their mission leverages both digital and physical innovation: As John Dean puts it, “no matter how smart AI is, you need that ground truth data,” making WindBorne’s approach durable for the long haul of climate adaptation and planetary-scale intelligence.
Call to Action: For listeners interested in joining, John underscores WindBorne’s talent needs, especially for those eager to work on mission-driven, global-impact technology at the boundary of AI and physical infrastructure.