Funding Brief｜Atomic Raises $12.5M Series A, Ex-Tesla Team Hands 90% of Purchasing to AI

🌏 中文版

Funding Details

What the company does

Atomic automates supply-chain purchasing decisions — using AI to decide how much inventory a company should hold and where, simulating scenarios and then recommending, or automatically choosing, how to respond as demand shifts.

All three founders came from Tesla's supply-chain planning team, and they built Atomic's core around the experience of handling supply-chain crises there: the system simulates multiple scenarios and recommends how to split orders between primary and backup suppliers, and complex logic changes can be built and applied to daily purchasing decisions within about an hour. Atomic's software now automates 90% of purchasing across hundreds of sites for DoorDash's DashMart business. At Good Chop, HelloFresh's meat-box subscription business, Atomic helped cut inventory on hand from eight or nine weeks down to four while revenue more than doubled and the distribution network and product range expanded at the same time.

Atomic came out of stealth in 2025 and has raised just north of $15M to date.

What this round signals

What it means for the agent ecosystem

Supply-chain decision automation is filling up fast with competitors: Lyric raised $43.5M for AI decision products across supply chains, Freehand raised $75M for agents that manage supply-chain spend at companies including Meta and Unilever, Didero raised $30M for procurement, Magentic raised $18M, Pelico raised $40M to cut factory delays, and Doss raised $55M for AI inventory management that plugs into ERP systems. Incumbents are moving too — Kinaxis acquired MPO for $45M to add real-time order execution. Atomic's angle is connecting planning logic directly to the order that actually goes out, rather than just producing recommendations.

What investors are betting on

Klass Capital founder Daniel Klass put it this way: "We've never heard customers talk about a software vendor the way they talk about Atomic. They trust it with daily operating decisions and are asking for more." That's a bet on depth of trust — once a customer is willing to hand a decision that directly hits cash flow, like how many units to order, to an agent, the switching cost and risk both become very high, creating natural stickiness. Klass Capital's own portfolio already includes supply-chain software company Nulogy, making this an add-on bet within the same category.

Numbers worth watching

• 90% of DashMart's purchasing is automated across hundreds of sites — one of the most heavily automated supply-chain agent deployments disclosed publicly to date
• Good Chop's inventory-on-hand fell from 8-9 weeks to 4 weeks while revenue more than doubled — one of the few cases with both efficiency and growth data to back it up
• An SEC filing shows the round actually raised $18.5M, about $6M more than the announced $12.5M, likely because the official figure only counts the first close

Watchlist status

Atomic is not yet on the watchlist. Recommend adding it to section D9 (Industrial / Manufacturing / Supply Chain), with tracking focus on: whether it can replicate beyond DashMart into more retail and food-service supply-chain customers, whether its "planning-to-order" integration holds up against recently funded rivals like Lyric, Freehand, and Didero, and whether the gap in the SEC-disclosed raise amount gets any further explanation.

Today's takeaway

At least six or seven companies in supply-chain automation have raised money recently, and Atomic's edge isn't model capability — it's connecting planning logic directly to the order itself. When a company is willing to let an agent decide whether to place an order rather than just suggest one, that's a trust threshold that's already been crossed.

References

• Exclusive: Ex-Tesla team raises $12.5M to put supply chains on autopilot
• Atomic Raises $12.5M to Build the AI Control System for Physical Goods Companies
• Ex-Tesla planners raise $12.5M from Klass Capital and Madrona to let AI place company orders