InMotion Ventures has participated in Matta’s $14 million raise. The seed round was led by Lakestar, with participation from Giant Ventures, InMotion Ventures, Redseed, 1st Kind, Unruly Capital, and Boost VC
Sam Nasrolahi, Principal at InMotion Ventures, explores why we invested
The market
Manufacturing underpins one-third of our global economic output. It is also one of the last major sectors to capture the benefits of applied AI. For all the progress in automation, factories remain reliant on manual inspection, intuition, and siloed systems to keep production moving.
Quality control in particular remains a stubborn bottleneck, it is slow, error-prone, and overly dependent on increasingly scarce human expertise. At the same time, manufacturers across all sectors are under pressure to drive efficiency, reduce waste, and meet rising standards for quality and sustainability.
There has been no shortage of new technologies promising to reimagine how products are designed and made, most have struggled to get out of first gear. The initial excitement around computer vision and automation tools faded when the realities of full-scale rollouts became clear. Deployments were slow, data-hungry, and expensive. Put simply, these novel technologies weren’t suited to high-variance, high-speed environments.
New advancements in industrial AI are creating a real opportunity to give machines the ability to see, understand, and respond to their environment in real time. Compute constraints are diminishing rapidly (significant improvements in edge hardware enable previously impossible low-latency applications), camera costs are plummeting, and software models continue to become more sophisticated. What’s missing is a platform that makes intelligence immediately deployable, at scale. One that bridges AI research with the heterogeneous reality of factory floors.
What does Matta do?
Matta builds sentient factories. Their full-stack system combines advanced AI models, cameras deployed at the edge, and a modular operating system that learns the rules of production and applies them directly on the line. It automates quality control and anomaly detection, performs measurements, diagnoses root causes, and recommends corrective actions in real time.
This is achieved through the ‘MattaBox’, a compact industrial unit that connects to standard cameras and performs all computation locally. The design avoids the latency and compliance drag of remote loops, allowing the system to observe a short period of production, form a baseline of “normal,” and begin flagging deviations immediately. It uses unsupervised and self-supervised learning to radically reduce (or eliminate) the need for labelled data.
With this approach, Matta’s system does not require the vast quantities of training that typically hold back deployments. Most installations are completed in hours, not months. In one polymer manufacturing deployment, Matta achieved over 99% defect-detection accuracy with just ten minutes of data.
Matta executes a two-pronged commercial strategy. First, they sell directly to manufacturers via a plug-and-play system that can retrofit into existing lines. Second, they partner with machine OEMs to natively embed their AI into next-generation equipment, ensuring Matta becomes the default intelligence layer for new production capacity entering the market.
Why is it different?
Industrial AI is a crowded field, but Matta’s approach stands out for three reasons:
- Full-stack simplicity. By integrating off-the-shelf hardware, edge-native software, and unsupervised computer vision, Matta can control the entire workflow from observation to image capture to insight. All of this leads to a total bill of materials roughly 10% the cost of legacy incumbents, rapid deployment and best-in-class reliability.
- Factory-first design. The system is purposely designed for the realities of high-variance factory floors. It is low-cost, high-speed, and capable of working across everything from electronics and automotive to defence and apparel.
- Scalable defensibility. Each installation generates proprietary datasets of visual and operational intelligence, compounding Matta’s advantage as more machines come online.
In practice, this makes Matta more like an operating layer for industrial perception, rather than another point solution.
Why did we invest?
The team. CEO, Douglas Brion, and CSO, Sebastian Pattinson, combine deep industrial and AI expertise from Cambridge, Imperial College, and MIT. They’ve assembled an elite team of engineers who understand the physics of manufacturing and the realities of deploying into complex environments. They move quickly, speak the language of factory engineers, and have already shown they can partner effectively with major manufacturers across a variety of sectors.
The technology. What really sets Matta apart is the flexibility of its AI. Unlike traditional computer vision systems that require extensive data labelling and retraining, Matta’s models learn by noticing differences. The system identifies anomalies and allows human operators to confirm whether the difference is noteworthy through a simple UI. Over time, this human-in-the-loop feedback educates the model, making it smarter, faster, and more attuned to the realities of each production line. The result is an inspection system that can be deployed in hours and adapts to new environments without retraining.
The timing. AI is coming to the physical world, and the factory floor is the next frontier. Manufacturers are realising that future productivity gains will come from intelligent systems that make sense of the data they already have. Matta enables that shift, turning cameras and machines from passive recorders into active participants in production.
We believe Matta represents the beginning of a new industrial paradigm, one where every machine can see, think, and optimise itself in real time. As factories evolve into sentient, self-improving systems, Matta’s technology will form a critical layer of that transformation.
We’re excited to partner with Douglas, Sebastian, and the team as they bring intelligence to the heart of manufacturing.