Boston supply-chain startup Atomic closed a $12.5 million Series A to push its AI planning software from recommendations toward autonomous purchasing decisions, TechCrunch reported on Sept. 29, 2026. Klass Capital and Madrona Venture Group led the round, bringing Atomic’s total funding to just over $15 million, according to that report and a same-day PR Newswire release.
Co-founders Michael Rossiter and Neal Suidan—former Tesla planning leaders—built early versions of the system during the 2018 Model 3 production ramp. Longtime Tesla planning director Jeff Goodrich has joined as CTO and third co-founder.
Atomic Series A: Customers and ARR Growth
TechCrunch reported that Atomic’s annual recurring revenue has quintupled since the start of 2026, citing Jon McNeill of DVx Ventures, where the company was incubated. Named customers include DoorDash and HelloFresh. McNeill said DoorDash runs roughly 90% of purchasing across hundreds of DashMart sites through Atomic’s platform.
Atomic’s Nucleus AI agent layer already handles tasks such as S&OP preparation, inventory questions and supply-risk checks for customers, the company said in its release. The new capital is earmarked to extend planning into a control system that connects business objectives to daily buy, build and move decisions—and to support larger enterprise deployments.
Boston’s Stake in Physical-Goods AI
Atomic’s pitch matters for New England manufacturers, CPG brands and food operators wrestling with spoilage, inventory swings and spreadsheet-bound planning. Rossiter told TechCrunch that AI can search an “infinite” optimization space that changes constantly—echoing Tesla’s emphasis on decision speed as a compounding advantage.
The raise sits alongside other recent Boston AI funding. For related local startup coverage, see our Harvey Boston office report and our overview of major AI companies in Massachusetts.
Our Take
Atomic is selling autonomy where many planners still live in Excel. If DoorDash-scale purchasing holds and onboarding stays fast, Boston gains a credible “AI for physical goods” operator that is not another chat wrapper. The risk is classic enterprise AI: customers may love recommendations but hesitate to let software place the orders. Closing that trust gap is the real product after the Series A.
Primary sources: TechCrunch on Atomic’s Series A; Atomic PR Newswire release; Atomic.