Noetive Raises $41M to Build Physical Operations AI

A factory can know where every pallet, machine, worker, and order sits, then still depend on one veteran operator to decide what happens when the plan breaks. Noetive has emerged from stealth with a $41M Seed round to turn more of that operating judgment into an AI system that works across software, people, and the physical environment.

Eclipse led the financing, with Craft Ventures, The Westly Group, Swish Ventures, Factory, Incite Ventures, Gigascale Capital, Operator Partners, and Liquid 2 Ventures participating. The San Francisco company was founded in 2026 by CEO Amir Frenkel and CPO Dan Barak, and it is already working with design partners in physical industries while building what it calls an "intelligence of record."

What Noetive Announced

Noetive announced the $41M Seed round on September 16, 2026. The round also included individual investors with operating backgrounds in AI and hardware, including Meta CTO Andrew Bosworth, Airbnb CTO Ahmad Al-Dahle, Nest co-founder Matt Rogers, Carnegie Mellon professor Yaser Sheikh, Gigascale Capital founder Mike Schroepfer, Bedrock Robotics co-founder and CEO Boris Sofman, former Founders Fund investor Zach Frenkel, and Decart co-founders Dean Leitersdorf, Moshe Shalev, and Orian Leitersdorf.

The company said the capital will support frontier research into self-improving AI for physical operations, field work with a curated set of design partners, and hiring across AI research and full-stack engineering. Noetive did not disclose its valuation, financing terms, ownership changes, investor allocations, revenue, pricing, customer count, or any prior financing, so the public record supports a $41M Seed event rather than a broader capital history.

How Noetive's Intelligence of Record Works

Most industrial businesses already have systems of record. Enterprise resource planning, warehouse management, transportation management, manufacturing software, sensors, spreadsheets, and paper logs can each describe part of an operation, but they do not necessarily explain how a decision in one area changes labor, inventory, machinery, delivery schedules, and customer commitments somewhere else.

Noetive's proposed intelligence layer combines agents that sit on top of existing business systems with multimodal sensing pods that observe conditions on the ground. The company says the system learns how an operation actually runs, absorbs working knowledge from operators, takes on end-to-end tasks, and uses the results of each decision to improve the next one. Its initial work includes dynamic production planning in manufacturing and manifest consolidation in logistics, with the broader website also naming construction as an early vertical.

That design puts the company's thesis beyond a conventional analytics dashboard. A dashboard can surface a delayed order or an idle line; Noetive is trying to connect the signals, choose a response, coordinate the handoffs, and adapt as conditions change. The commercial burden follows directly from that ambition because customers will need the system to understand local exceptions, produce measurable gains, and behave reliably when its decisions touch real inventory, equipment, schedules, and people.

The Early Evidence and Its Limits

Noetive has named Steuben Foods as an early design partner. In the launch announcement, Steuben Foods CEO Menachem Katz said production planning that previously happened monthly and took a week can now happen daily in minutes because Noetive combines system data with conditions on the factory floor.

That is a specific operating claim, but it remains a company and customer-reported example rather than an independently audited performance study. Noetive has not published a customer count, deployment count, retention data, pricing, measured savings, error rates, safety record, or a comparative benchmark. The result is useful evidence of where the product enters an operation, while the scale and repeatability of the improvement remain open questions.

Why Eclipse Built Before It Invested

Noetive also reflects a different kind of seed formation. Eclipse says it built the company from the ground up with Frenkel after he joined the firm as its first chief AI officer in April 2026. The venture firm helped shape the thesis, assemble the founding team, and connect the company with physical-industry operating problems before leading the financing.

Frenkel spent nearly a decade as a vice president at Meta and previously held leadership roles at Alphabet and Amazon. Barak, Noetive's co-founder and CPO, previously served as vice president of product at Netlify after co-founding Stackbit. Noetive's current team page lists researchers, engineers, and operators with backgrounds across Meta, Google, Amazon, and Fortune 500 companies, but it does not identify a current CTO.

The company-building model matters because industrial AI needs more than model access. The first product decision is often choosing a problem that is painful now, repeatable across customers, measurable in economic terms, and rich enough to teach the system something reusable. Eclipse says Noetive selected production planning and manifest consolidation against those criteria, using deployments to move from solving a narrow task toward a broader platform.

What the $41M Must Prove

A $41M Seed round gives Noetive room to recruit scarce technical talent and stay close to difficult customer deployments. It also raises the standard of proof. Self-improving agents and sensing hardware must survive changing plant conditions, incomplete data, tacit operator knowledge, integration constraints, and the consequences of acting on a bad plan.

The strongest part of Noetive's launch is the choice to begin inside live operations rather than treating industrial work as a clean software abstraction. The risk lives in the same place. Every factory, logistics network, and construction site carries its own machinery, workflows, economics, and exceptions, so the company must show that its learning compounds across deployments instead of becoming a series of expensive custom projects.

Noetive is financing the handoff from an embedded operating experiment to a repeatable system of judgment. Production planning and manifest consolidation are the first tests, and the next evidence will come from whether the system can keep earning more responsibility after the floor changes, the order moves, and the experienced operator is no longer the only person who knows what to do.