There is a moment in every market when the noise drops out and the signal finally speaks. Time does that. It tells on everyone. Revenue curves. Load spikes. Supply chains coughing at 2 a.m. Most companies stare at the clock. Nixtla decided to teach it how to talk back.
Nixtla just closed a $16M Series A, led by Energize Capital, with True Ventures and GreatPoint Ventures leaning back in like they knew the second verse already slapped. San Francisco energy. Global consequences. This is what happens when timing stops being a guess and becomes an asset you can ship.
Founded in 2021 by builders who speak applied mathematics and machine learning the way jazz musicians hear keys, Nixtla grew up in the open. 45M+ downloads across the Nixtlaverse. 15K GitHub stars. A community that stress tested the code before the suits ever showed up. Open source does not mean soft. It means exposed to reality early.
At the center is TimeGPT, the first production-ready foundation model built specifically for time series. Not language pretending to understand numbers. Not dashboards with opinions. 100B+ data points across retail, energy, finance, and IoT, trained to see patterns where humans see spreadsheets. The result is forecasts that land 42% more accurately with 10x better inference efficiency. Time stops wasting cycles.
Enterprises noticed. Microsoft runs sub 5-second deployments on Azure AI. Lyft cut false alerts by 85%. Global brands like Zalando, Decathlon, Prudential, and Unilever did not show up for the demo. They stayed for the math. When forecasting stops lying, operations start telling the truth.
This round puts more fuel behind agentic forecasting, enterprise scale deployments, and a platform that treats time as a first class citizen, not an afterthought. Nixtla Enterprise 2.0. TimeGPT 2.1. Multivariate models that understand the argument between variables instead of averaging them into silence.
Credit where it belongs. Max Mergenthaler Canseco, CEO & Co-Founder, has been steady on the cadence since day one. Cristian Challu, Ph.D., CSO & Co-Founder, keeps the science honest. Han Wang, CTO, brings production scars from Lyft, Tecton, Amazon, and Quantlab, the kind you earn when models meet traffic. This team does not romanticize complexity. They domesticate it.
Energize Capital adds more than capital with Juan Muldoon, Partner, joining the board, a signal that infrastructure-grade thinking belongs in the room when forecasting touches energy grids, factories, hospitals, and balance sheets. True Ventures and GreatPoint Ventures returning says the earlier math checked out.
Here is the quiet lesson. Nixtla did not chase trends. They trained on time itself. They built where decisions actually break. Forecasting is not about predicting the future. It is about reducing regret in the present. When time finally speaks clearly, the smart companies listen, and the rest keep checking the clock, wondering why it keeps winning.
