Foundry Robotics has raised an oversubscribed $100 million Series A round to expand its autonomous manufacturing operations and develop AI-powered robotic assembly for complex industrial products. Led by Marlinspike Partners, the round includes existing investors Khosla Ventures and Red Glass Ventures, alongside new participants SE Ventures, Wing VC and Axiom Ventures. The company plans to expand its 130,000-square-foot Emeryville facility, deploy more AI models on factory floors and grow its robotics, engineering and manufacturing teams.
Foundry Robotics Raises $100M to Build Flexible AI Factories
Manufacturers are under pressure to produce complex hardware faster, particularly in sectors such as defense, robotics and data center infrastructure. But while product design cycles are accelerating, physical assembly remains difficult to scale. Traditional production lines can take months to deploy and are often expensive to adapt when products change.
Foundry Robotics wants to address that constraint with autonomous factories built around AI-powered robotic assembly cells. The company has raised an oversubscribed $100 million Series A round led by Marlinspike Partners, with participation from Khosla Ventures, Red Glass Ventures, SE Ventures, Wing VC, Axiom Ventures, Hanabi and Zero Shot Fund.
The funding will support expansion of Foundry’s Emeryville, California, facility, accelerate the deployment of AI models in manufacturing and expand its research, engineering and factory operations teams.
A Factory Designed Around Robotics
Foundry builds and operates factories that combine robotic systems with experienced manufacturing specialists. Rather than adapting conventional production lines for automation, the company is designing its facilities around robotic deployment from the outset.
Its manufacturing-as-a-service model allows customers to use Foundry’s production capacity without building their own dedicated assembly lines or investing in expensive retooling. The company says it can move customers from design to production in weeks, although the announcement does not specify whether that timeline applies across all product types or manufacturing complexity levels.
Foundry has expanded its manufacturing footprint twentyfold to a 130,000-square-foot facility in Emeryville. Its target applications include drones, autonomous vehicle subassemblies, wire harnesses, underwater vehicles, telecom and data center infrastructure, ground robots, rugged computers and satellites.
These products often combine electrical, mechanical and other components that require precise assembly. Such work can be difficult to automate using fixed production systems designed around a single product.
Foundry’s approach instead centers on flexible robotic cells intended to accommodate different assembly tasks. That flexibility could help manufacturers respond to changing demand without repeatedly rebuilding production lines, though the extent of product changeover capability will depend on the specific task and hardware involved.
Physical AI Meets Manufacturing Operations
At the core of Foundry’s model is the use of AI models trained on the building blocks of assembly and deployed across a network of robotic cells.
The company says its FoundryOS platform connects the robots, factory-floor operations and cloud infrastructure through a unified agentic system. Production runs feed information back into the models, allowing the system to improve its performance over time.
In principle, this approach could let manufacturing knowledge gained from one assembly task inform related tasks in the same facility. Foundry gives the example of a robotic cell assembling a drone one day and a satellite component another day, with experience from production helping improve both processes.
The practical challenge is transferring learned capabilities between products with different geometries, components, tolerances and assembly sequences. Manufacturing-grade precision also requires reliable hardware, quality controls and mechanisms for handling failures or unexpected conditions.
Foundry’s announcement does not provide independent benchmarks for throughput, defect rates, production costs or how quickly its cells can be reconfigured. Those measures will be important in determining how its approach compares with conventional automation and other flexible manufacturing systems.
Defense and Industrial Demand Drive Expansion
Foundry serves both defense and commercial customers, with an emphasis on high-value products where limited production capacity can constrain delivery.
The company sees particular demand for additional domestic assembly capacity in unmanned systems and other industrial hardware. HII Unmanned Systems President Duane Fotheringham described production capacity as a constraint on getting unmanned systems to military users and pointed to Foundry’s approach as a potential capability for the defense industrial base.
The broader opportunity extends beyond defense. Data center infrastructure, robotics and autonomous systems all depend on specialized hardware that must be assembled reliably at scale. A flexible manufacturing model could be useful where demand changes quickly or product designs evolve faster than conventional factory layouts can accommodate.
However, scaling autonomous assembly across different product categories is not simply a matter of adding robots. Each application can require new tooling, process validation, quality assurance and integration with customer requirements. Foundry’s ability to standardize these elements while supporting varied products will be central to its expansion.
Funding Supports the Next Stage of Deployment
Foundry was co-founded by Adarsh Kulkarni and Emmanuel Lamacchia. Kulkarni previously worked at Ghost Robotics and Scale AI, including robotics-related roles, while Lamacchia was a program manager for Tesla’s Model Y program. The company has also added personnel with backgrounds in software, robotics and manufacturing.
The new capital gives Foundry resources to expand its factory footprint and further develop its AI-driven production model. Its manufacturing-as-a-service approach could reduce the need for customers to build dedicated facilities, but its commercial impact will depend on production economics, reliability and the range of products its systems can handle.
For now, Foundry is betting that the next generation of manufacturing will combine physical AI with factories designed specifically for flexible automation. If the company can demonstrate that its robotic cells adapt efficiently while maintaining production quality, it could offer manufacturers another way to expand assembly capacity without repeatedly rebuilding their production lines.
Market Landscape
AI-powered manufacturing sits at the intersection of robotics, industrial automation and machine learning. The opportunity is to make production systems more adaptable without sacrificing reliability or precision.
- Physical AI: AI models interact with real-world environments through robots and other physical systems.
- Flexible automation: Robotic cells are designed to support multiple tasks rather than a single fixed production line.
- Manufacturing-as-a-service: Customers access production capacity without necessarily owning and operating their own factories.
- Defense industrial capacity: Autonomous systems and other specialized hardware create demand for scalable assembly.
- Factory software: Platforms connect robotic hardware, production workflows and data to coordinate operations and support improvement.
Foundry’s differentiation rests on combining factory ownership, robotic cells and its FoundryOS software. Its longer-term position will depend on whether it can deliver repeatable production quality and competitive economics across varied assembly tasks.
Top Insights
- $100 million Series A: Foundry Robotics secured an oversubscribed round led by Marlinspike Partners.
- 130,000-square-foot facility: The company says it has expanded its Emeryville manufacturing footprint twentyfold.
- Flexible robotic assembly: AI-powered cells are designed to support complex products across defense and commercial industries.
- FoundryOS integration: The platform connects robots, factory operations and cloud infrastructure in an agentic system.
- Manufacturing-as-a-service: Foundry operates the factory, aiming to help customers move from design to production without building dedicated assembly lines.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
