Strive Robotics Raises $3M+ for Adaptable Factory Robots

Strive Robotics wants the next piece of factory automation to learn its way around equipment the manufacturer already owns. For a shop whose parts and production requirements keep changing, that choice puts the work of adapting automation inside the buying decision from the start.

The Cambridge, Massachusetts startup announced more than $3M in pre-seed funding on November 5, 2026. Hi Ventures led the round, with Grid Capital, Tenon Ventures and Thin Line Capital participating. Co-founders Craig Cheney, CEO, Geehoon Park, CTO, and Paul Guiloff, CRO, are developing physical AI systems for high-mix manufacturing, using the financing for product development, hiring and deployments.

The commercial idea reaches beyond getting a robot to repeat a movement. Strive wants manufacturing teams to teach and reconfigure production workflows themselves, giving the people who understand the job more control over the automation expected to perform it. That makes their existing process knowledge part of what the company needs to deliver its product.

What Strive Robotics' pre-seed funding supports

Strive's announcement describes a progression from prototypes toward sustained operation in manufacturing environments. The money is intended to support the engineering and deployment work along that route, rather than a disclosed expansion of an already quantified installed fleet. The company has not published customer revenue, measured production gains or a valuation in the reviewed primary materials.

The company identifies Hi Ventures as the lead investor. Its own LinkedIn announcement names Grid Capital, Tenon Ventures and Thin Line Capital alongside the lead, providing the full identity behind the release's shortened reference to Thin Capital. The announced amount remains more than $3M; the primary announcement does not specify the dollar denomination or the financing's closing date.

Strive is targeting the work around industrial equipment, including material handling, fixture preparation, loading and unloading, operating controls and checking results. A machine can perform its own production function while still needing someone to arrange the work before and after the cycle. Strive's opportunity sits in that surrounding sequence, particularly when the next assignment differs enough to make a fixed setup awkward to reuse.

That scope gives the founders a demanding product definition. A buyer interested in automation has to understand which steps the system can own, where a person remains involved and how the workflow changes when another part arrives. The financing provides resources to develop those answers with manufacturers already doing the work.

Existing machine controls become part of the product

Strive describes robots that work with familiar machine interfaces, including buttons, touchscreens and keyboards. It also describes interchangeable end-effectors designed for robots. Those choices allow the company to approach a factory through the equipment and operating methods already present, while selecting tools suited to the robot's job.

This is a useful distinction in how a manufacturer evaluates change. Replacing or redesigning a functioning production environment introduces work beyond the robot purchase itself. Building around existing equipment can preserve more of that investment, although each application still needs its own assessment of fit, reliability and safe operation. Strive has not published evidence establishing universal compatibility with factory machines.

The company's robotics engineering description places classical controls, model-based perception and fine-tuned vision-language-action models inside its approach. That combination matters because the commercial assignment includes physical behavior and recovery, alongside interpreting what a person asks the system to do. A useful manufacturing workflow has to survive its interactions with parts, sensors, tools and the surrounding equipment.

Strive calls its teaching approach Apprentice Mode. It is developing ways to learn workflows through demonstrations and instructions, with the aim of reducing the need to program each task from scratch. The economic attraction is the possibility of carrying automation into another assignment with less outside engineering, a particularly relevant proposition when production changes frequently enough to become part of an ordinary working week.

The operator's knowledge has a new destination

Strive's leadership team combines robotics research and commercial company-building. Craig Cheney is an MIT SB, MS and PhD graduate whose background includes the Robotics & AI Institute. Geehoon Park holds an MIT robotics PhD and previously worked at Boston Dynamics, while Paul Guiloff is an MIT MBA and engineer responsible for strategy, customer discovery and go-to-market, according to the company.

Their product puts the manufacturing operator close to the teaching process. Strive describes work instructions, CAD and factory data as inputs, followed by execution, adaptation and verification. The person who understands the sequence can supply information that would otherwise have to be discovered and translated during an integration project.

The homepage says Strive is working with a high-mix manufacturer on an end-to-end workflow using real production parts and equipment. That is company-reported fieldwork with an unnamed manufacturer; the reviewed materials do not establish a deployment count, customer return on investment or independently measured reliability. The practical work under development includes kit preparation, loading and unloading, and fixture or parts assembly.

For the operator, being able to teach a job would create another way to use accumulated process knowledge. For the startup, making that teaching usable means accepting responsibility for the machinery between an instruction and a checked physical result. The quality of that transfer affects how much the manufacturer can do without bringing a separate robotics specialist back into each change.

Reconfiguration is a manufacturing buying decision

The integration problem has a longer history than the current physical AI label. NIST's work on emerging robotics technologies identifies selection, installation, operating robustness and rapid retasking as challenges for small and medium manufacturers. NIST also emphasizes evaluation methods for technologies intended to work in high-mix, low-volume environments.

That context helps locate Strive among the different commercial routes into industrial robotics. The relevant comparison includes conventional fixed automation and integration-led deployments, as well as more adaptable systems. A manufacturer choosing between them needs to consider the frequency of job changes and the people required to support those changes, alongside the initial machine capability.

DevCuration's recent coverage includes Aformic's factory robotics expansion and Formic's factory robotics financing. Those records offer related paths into the factory for readers following the sector, with their own product scope and capital structures.

Strive's careers page lists Boston-based roles spanning robot learning, perception and robotics engineering. The hiring priorities place more engineering capacity around the teaching, sensing and physical execution described in the product, while deployments bring those disciplines into contact with a manufacturer's actual sequence of work.

For a factory evaluating Strive, the next conversation can begin with an existing job: its parts, fixtures, controls, instructions and the points where an experienced person intervenes. Those details give the founders a specific assignment for the new capital and give the manufacturer a way to judge how much of its next setup the automation can carry forward.