A New Entrant Targets Construction's Productivity Gap
Construction is one of the last major industries to resist the automation wave. Over 41% of the U.S. construction workforce is expected to retire by 2031, and the sector has seen a 46% decline in productivity while nearly every other industry has posted gains from AI and software. Outdoor jobsites — exposed, muddy, variable in weather and layout — are simply harder to automate than factory floors.
Gritt, a San Francisco-based startup launching publicly on July 21, 2026, believes it has the answer: physical AI that rides on equipment crews already own and maintain.
$32.4M in Funding
Gritt closed $32.4 million across pre-seed and Series A rounds:
- $26M Series A led by Obvious Ventures, with participation from Union Square Ventures and Active Impact Investments
- Previous investors including First Round Capital, Climactic, Congruent Ventures, and VSC Ventures also joined
“Every industry has embraced automation, but outdoor construction sites are exposed, with unforgiving terrain and variable weather. You can't tell a construction site to be less muddy or to stop snowing,” said Puneet Puri, CEO and co-founder of Gritt. “Gritt's physical AI is designed to work in this unstructured, outdoor world.”
How Gritt Works
The core insight is pragmatic: rather than replacing the heavy equipment that construction crews already know, Gritt's robotic arms and AI systems attach to existing jobsite machinery — skid steers, forklifts, and other common platforms — turning them into semi-autonomous workstations.
The system performs:
- Pick-and-place operations with millimeter precision
- Assembly tasks across changing site layouts
- Material transport on active construction sites
Every deployment feeds data back into the intelligence layer, so tasks that took months to train the first time can now be adapted to new sites in days. Gritt also provides site supervisors with real-time data on completed work, material movement, and site conditions to help with planning and decision-making.
Proven in Solar and Infrastructure
Gritt's systems are already deployed at scale on large-scale solar farm construction sites, working with top-10 U.S. construction companies. The company claims it has achieved autonomous placement of tens of thousands of solar panels with zero breakages — an impressive metric in an industry where even minor breakage rates can eat into margins.
Andrew Beebe, Managing Director at Obvious Ventures, said: “Every wave of automation started with someone willing to go where the work was hardest and least predictable. Gritt is bringing physical AI to the toughest construction sites in the world. Every site they deploy builds the flywheel that makes the system smarter and more capable.”
The Founding Team
Gritt was founded by robotics and AI experts from Carnegie Mellon, Stanford, and MIT:
- Puneet Puri — CEO and co-founder
- Vishal Dugar — CTO and co-founder
“Nobody is solving the repetitive, high-volume manual work that shapes the world's infrastructure,” said Dugar. “We need an intelligence layer that can complete dexterous tasks and make decisions in an environment that changes every hour. This is one of the most challenging frontiers in physical AI.”
What It Signals
Gritt's launch is notable for three reasons:
- Physical AI is moving indoors → outdoors. Most robotics investment has targeted factories and warehouses. Gritt is part of a newer cohort tackling unstructured, outdoor environments.
- The attach-to-existing-equipment model. This dramatically reduces customer adoption friction and capital cost, and is a pattern we may see more of in heavy-industry robotics.
- Top-tier investor backing. Obvious Ventures, Union Square Ventures, and First Round Capital are all credible names that rarely invest in capital-heavy robotics — suggesting they see software-style margins in Gritt's intelligence layer.
With the infrastructure buildout boom continuing — solar arrays, data centers, bridges, roads — demand for construction automation is only going to grow. Gritt enters the market at a moment when the labor shortage is acute, the technology is finally good enough, and investors are willing to bet on physical AI beyond the factory floor.
