Alloy Robotics has raised $8 million in a funding round to develop AI agents that help engineers identify failures and performance issues across robot fleets, as robotics companies increasingly rely on large volumes of operational data to maintain machines in the field.
The round values the company at $80 million, just over a year after its founding. Square Peg led the financing, with existing pre-seed investors Blackbird, Airtree and Skip Capital participating again. The round also attracted executives and engineers from OpenAI, Anthropic, Tesla, Waymo, Halter and Carbon Robotics, alongside several Alloy customers. Forbes first reported the financing.
Alloy’s software is designed to bring together data that is typically spread across different systems, including fleet logs, telemetry, video, sensor readings and engineering records from platforms such as Slack and Jira. Its AI agents analyse the information for anomalies, regressions and recurring failure patterns, while linking findings to the missions, timestamps and underlying signals involved.
The company says the approach is intended to reduce the amount of manual investigation required when robots malfunction. Engineers can spend days or weeks tracing a failure through disconnected datasets, particularly as fleets become larger and operate across varied environments.
At navigation technology company Advanced Navigation, Alloy has reduced field-test analysis from about a day to less than 10 minutes, according to the company. Product validation manager Jai Castle said the faster analysis allowed the team to shift its focus from completing existing tests to pursuing additional testing. In one period, the team completed 44 field tests in slightly more than a day.
The platform has also been used by U.S. autonomous-drone startup DroneForge. Engineer David Crabtree said Alloy helped challenge an initial diagnosis of a component failure by showing that both state estimators were operating normally and pointing engineers towards the actual fault.
Alloy currently supports close to 1,000 robots and has analysed more than 10,000 missions, with most of that activity occurring during the past two months, according to the company. Its customer base spans navigation, defence, drones, agriculture, maritime systems, humanoid robots, construction and medical robotics.
The company is also integrating its platform with AI coding tools. Through a native Model Context Protocol (MCP) server, Alloy allows coding agents including Codex and Claude Code to access mission-level context. Engineers can therefore query operational data and investigate failures without manually collecting files from separate systems.
The development comes as robotics companies move from controlled testing environments towards larger commercial fleets. As deployments expand, identifying whether a problem is caused by hardware, software, environmental conditions or changes in system behaviour becomes increasingly important. Historical fleet data can also provide a reference for identifying recurring failures and detecting regressions before they spread across larger deployments.
Joe Harris, founder and CEO of Alloy Robotics, previously served as chief commercial officer at Eucalyptus before its acquisition for $1 billion. He said the company’s focus is on extracting useful evidence from the large amounts of data generated by robots.
The new funding will be used to expand Alloy’s engineering team, grow its presence in the U.S. and further develop its AI models and agent platform.
The company is positioning the technology around a broader challenge in robotics: making fleets more reliable as they scale. Rather than treating each failure as an isolated engineering problem, Alloy’s system is designed to turn operational experience across multiple missions into information that can be used to diagnose subsequent problems and improve future deployments.
