A week in AI is like a year in other industries. I hope these issues become your weekly source of AI information, inspiration, and ideas.
Good morning. Some of the most interesting AI companies don’t look like AI companies at all.
This one started with wildfires, a frustrated wife, and a founder who decided to actually fix the problem.
He built a better fire nozzle.
Now he might be sitting on one of the most valuable datasets in AI.
Let’s dive in.
🤝 This edition is brought to you by Airia
Ready to revolutionize your enterprise AI?
With Airia, you gain seamless AI management, from deployment to governance.
Overcome adoption barriers and maximize efficiency without compromising security or scalability.
Airia empowers innovation, makes AI actionable, and helps your business achieve more with less.
It’s time to redefine enterprise AI on your terms.
Company background: HEN Technologies
Founded: June 2020
Team size: ~70
Funding to date: $30M+
ARR: $5.2M (2025)
Growth metric: 10x revenue growth in under 3 years, scaling through government and enterprise contracts
Sunny Sethi, HEN Technologies
Sunny Sethi does not fit the typical AI founder mold.
He is not building a coding copilot. He is not wrapping a frontier model. He is not talking about reinventing search or productivity.
He is building firefighting equipment.
That sounds niche until you look closer.
Sethi’s background spans nanotech, solar, semiconductors, and automotive systems. He had spent years solving technical problems across industries when a series of California wildfires hit much closer to home. The real turning point came in 2019, when he was away traveling and his wife was home alone with their young daughter during evacuation warnings.
She was furious. And she gave him a challenge that changed everything.
Fix it.
In 2020, Sethi founded HEN Technologies, short for High Efficiency Nozzles, with a straightforward mission: build firefighting tools that actually work better in the conditions people now face.
The first product was a nozzle designed to suppress fires faster while using far less water. According to the company, it can put out fires up to three times faster than older products while conserving roughly two-thirds of the water.
That alone would be a strong company.
But it is not the whole company.
The real product is not the nozzle
The nozzle is what gets HEN into the fire truck.
What matters next is everything the equipment sees, measures, and learns.
HEN has expanded beyond nozzles into monitors, valves, sprinklers, pressure devices, and flow-control systems. These are not dumb pieces of hardware. They are loaded with sensors, onboard compute, and custom circuit boards that track what is happening in the field in real time.
That means every fire response creates data.
How much water was used.
What pressure was required.
Which hydrant was tapped.
How weather conditions changed the response.
How flow rate behaved under live fire conditions.
Where equipment was located and how it performed.
This is what makes the story interesting.
Most companies would stop at “smart equipment.” Sethi kept going.
HEN built a cloud platform on top of the hardware, turning field data into usable intelligence for captains, battalion chiefs, and incident commanders. The system can help teams understand water usage, monitor active conditions, and eventually anticipate operational risks before they spiral.
In other words, HEN is building the digital nervous system for firefighting.
Why this matters for AI
This is where the story stops being a hardware story and starts becoming an AI story.
A lot of today’s AI is built on text, images, code, and synthetic environments. But the next generation of AI, especially robotics, predictive simulation, and world models, will need something harder to get: real-world physical data.
Not clean benchmark data.
Messy, high-stakes, real-environment data.
How does water behave under pressure in a chaotic environment? How does wind alter suppression? What happens when multiple systems draw from the same source? What changes when terrain, materials, or weather shift?
You cannot fully simulate that from a laptop.
You need sensors in the field. You need repeated deployment. You need real outcomes.
That is what HEN is quietly collecting.
Every piece of connected equipment expands the dataset. Every deployment improves the intelligence layer. Every fire response makes the system more valuable.
So while everyone else is talking about who has the best model, HEN is building something much harder to replicate: proprietary data from the physical world.
That is the kind of thing people will pay a lot for later.
Building where the stakes are very real
There is also something refreshing about how unsexy this all is.
Sethi is not selling into consumers. He is not shipping a trendy AI tool to early adopters. He is operating in one of the hardest markets possible, where trust matters, procurement is painfully slow, and getting it wrong is not an option.
Fire departments do not buy because something looks impressive in a demo.
They buy because it works.
And the buying process itself is tricky. HEN has to win over the people using the equipment on the ground while also navigating long government purchasing cycles. Sethi described the company as sitting in a strange middle: the product has to resonate like a consumer product, but the sales motion looks far more like enterprise and government infrastructure.
That is difficult to crack.
HEN seems to have cracked it.
The company launched its first products in 2023, generating $200,000 in revenue from 10 fire department customers. Revenue grew to $1.6M in 2024, then to $5.2M in 2025. This year, the company projects $20M in revenue.
It now serves 1,500 fire department customers, ships to 22 countries, and works with organizations including the Marine Corps, U.S. Army bases, NASA, and Abu Dhabi Civil Defense.
The bottleneck is no longer demand.
It is scale.
The playbook here is smarter than it looks
One thing I found especially interesting about HEN is how it approached the market.
The company did not begin by pitching a giant AI vision to the world.
It solved a painful operational problem first.
That gave it credibility. It created a wedge. It got the hardware into environments where data could start flowing. Then it layered intelligence on top.
That is a very different route from the usual AI startup strategy of model first, use case second.
HEN did the opposite.
Problem first. Infrastructure second. Data third. AI upside fourth.
That sequence matters.
Because once you own the infrastructure, the data starts compounding. And once the data compounds, the business gets much more interesting.
What comes next
Sethi says HEN’s long-term vision is not just better hardware. It is a predictive emergency response platform.
That could mean systems that warn teams when the wind is about to shift. Or when one truck is about to run low on water. Or when pressure changes across a shared hydrant network put an operation at risk.
It could also mean far more than firefighting.
If HEN keeps collecting high-quality physical world data, it may end up with something valuable to companies building robotics systems, predictive infrastructure, simulation engines, and world models.
That is the part to watch.
Because what looks like a public safety company today may turn into something much bigger tomorrow.
Takeaways
The best AI companies may not present as AI companies
Some are showing up through infrastructure, hardware, and category-specific systems rather than flashy consumer interfaces.
Own the data layer and the rest gets stronger
HEN did not start with a model. It started with a real-world problem and built the rails that generate proprietary data.
Boring industries can hide massive AI upside
The less crowded the category, the easier it can be to build something defensible.
Solving a real pain point buys you the right to expand
Fire nozzles opened the door. Data and intelligence became the bigger opportunity.
Physical AI needs physical data
That sounds obvious, but a lot of the market still acts like everything can be trained through simulation and scraped content. It cannot.
How's the depth of today's edition?
If one of these stories stuck with you, I’d love to hear which one.
Speak soon,
Lavena
P.S. If you want to get a founder feature about your own story, reply to this email. If you’d like to reach our newsletter audience (founders, creators, and marketers), click the button below.
If you’re new here, I’m over the moon you’ve joined us! To help me craft content that’s actually useful (and not just noise in your inbox), I’d love it if you took 1 minute to answer this quick survey below. Your insights help shape everything I write.
✨ Insane Media is more than one voice
Dive into our other newsletters - where psychology meets the creator economy, e-commerce marketing, and Human resources.
