Technology

Gritt Exits Stealth With $34M to Put AI-Controlled Robots on Solar Construction Sites

Martin HollowayPublished 2w ago4 min readBased on 1 source
Reading level
Gritt Exits Stealth With $34M to Put AI-Controlled Robots on Solar Construction Sites

Gritt, a robotics startup founded by Carnegie Mellon-trained roboticists CEO Puneet Puri and CTO Vishal Dugar, exited stealth on July 21, 2026 with $34 million in total funding and two systems already deployed in the field. The company's $26 million Series A was led by Obvious Ventures, with participation from Union Square Ventures and Active Impact Investment, building on an earlier seed round backed by First Round Capital, Climactic, Congruent Ventures, and VSC Ventures. TechCrunch

Gritt's approach is notable for what it chooses not to build. Rather than designing proprietary robots, the company rents off-the-shelf hardware — skidders and robotic arms from manufacturers like Kawasaki — and controls that equipment with its own AI models. The bet is that the bottleneck in industrial robotics lies in software and perception, not in the kinematics of the manipulator. That stance aligns with a broader industry pattern in which the cost of commodity manipulators keeps falling while the difficulty of making them useful in unstructured environments remains the real gating factor.

The first application is solar farm construction. Gritt's systems unload large glass solar panels, carry them to metal frames, and position them with sub-millimeter accuracy so human workers can fasten them in place. A typical eight-person crew installs about 800 panels per day. The same crew working alongside Gritt's systems can install 3,000 to 4,000 panels per day, according to the company. Two systems are currently operating in the field, and Gritt plans to scale to 48 deployed systems within the next six months.

The pipeline underpinning that ramp is substantial. Gritt is contracted to help install 2.8 gigawatts of solar panels over the next 18 months. Its customers include three of the top 10 US power construction companies. Andrew Beebe, partner at Obvious Ventures, led the Series A round, and the investor roster — blending climate-focused funds (Climactic, Congruent, Active Impact) with mainstream venture names (First Round, Union Square Ventures) — suggests the company is being framed as both a climate infrastructure play and a general-purpose robotics bet.

Gritt is not alone in targeting solar installation automation. Competitors include Luminous Robotics, Cosmic, and China's Trinabot, each pursuing variations on the problem of placing heavy, fragile panels at scale. The competitive landscape is early, with no dominant design or clear market leader established.

Gritt's positioning — "robots to build solar plants, then everything else," as TechCrunch characterized it — telegraphs an ambition that extends well beyond the initial solar use case. The hardware-agnostic, software-first architecture is, in principle, transferable to any construction or industrial task where heavy manipulation in semi-structured environments is the bottleneck. Whether the company can generalize its perception and control stack beyond solar panel placement, where the task parameters are relatively well-constrained, is the open question that determines whether Gritt becomes a vertical specialist or a horizontal platform.

The deployment metrics are the strongest signal in the announcement. Two fielded systems and a 2.8 GW contracted backlog give Gritt more validation than most robotics startups can point to at this stage. The 5x throughput improvement figure, if it holds across the planned 48-system fleet, would meaningfully change the unit economics of utility-scale solar construction — a sector where labor availability and installation speed are persistent constraints on project timelines.

The risk worth tracking is execution. Scaling from two systems to 48 in six months is an aggressive operational ramp by any measure, and field robotics has a long history of systems that perform well in controlled demonstrations hitting friction when deployed across varied sites, weather conditions, and panel geometries. The 2.8 GW backlog is contracted work, but contracted scope and delivered scope are different things. How many of those 48 systems are generating the claimed throughput figures in real field conditions by early 2027 will be the metric that matters.

For now, Gritt has arrived with a credible technical thesis, meaningful early traction, and enough capital to test its scaling hypothesis. The solar construction market needs exactly what the company is building — more panel-installation throughput per crew hour — and the competitive field is still wide enough that no incumbent has locked the category.