Notus Systems —

We'll be honest: Notus Autonomous Systems caught our attention not because of flashy marketing or a polished website (they barely have one), but because of what they're quietly building in San Francisco. This Y Combinator Spring 2025 company is tackling one of the most complex challenges in modern warfare: how do you control thousands of drones and robots without needing thousands of operators?

The Scale Problem They're Solving

Here's what impressed us most about Notus. While other defense tech companies focus on building better individual drones or robots, founders Arhum Jain and Emmanuel Margolin are thinking bigger. Their PALLAS system aims to flip the script from "one operator, one drone" to "one commander, thousands of devices."

That's not just ambitious, it's necessary. Modern conflicts have shown that drone swarms and mixed autonomous fleets are the future, but current command systems can't scale to match that reality.

Built for Real Combat Conditions

What sets Notus apart is their focus on contested environments where communications get jammed and GPS signals disappear. Most drone control systems fall apart when the enemy starts electronic warfare. PALLAS is designed to keep working even when things get messy.

The company says they're incorporating "combat user feedback and lessons from Ukraine," which suggests they're not building in a Silicon Valley bubble. They understand that battlefield conditions break a lot of assumptions about how technology should work.

Early Days, Big Potential

We have to be upfront: Notus is extremely early stage. Their website is basically a contact form, and they haven't announced any major funding rounds or customer wins yet. But sometimes the most interesting companies are the ones that focus on building rather than marketing.

The YC backing gives them credibility and resources to execute on their vision. If they can deliver on the promise of scaling drone operations from hundreds to thousands, they'll be solving a problem that every modern military faces.