Before a factory commits to manufacturing millions of chips, engineers stress-test the design in the fab. Before a product ships, a fault on the assembly line has to be caught before it becomes a recall. When AI performs this kind of work, it's referred to as physical AI: intelligence embedded in the equipment and engineering processes that produce everyday products.
Anthropic is partnering with UST, a technology and engineering services company that builds and operates the engineering environments its clients rely on to bring chips, cars, and connected devices to market. UST is deploying Claude inside those environments and training 20,000 of its engineers, architects, and consultants on Claude worldwide.
How UST Is Deploying Claude in the Production Processes Behind Physical Products
UST works alongside semiconductor, automotive, manufacturing, telecom, embedded, and IoT companies. It builds the systems those companies use to verify designs, validate chips, run factories, and service products once they're in the field.
These are long, multi-step processes where an early mistake grows more expensive with every subsequent step. A design flaw caught during verification costs an engineer an afternoon; the same flaw caught after a factory has committed to manufacturing costs an entire production run.
UST is integrating Claude into this work. Claude Code reads the schematics and pinouts an engineer works from, then writes and runs the tests that check the design. It carries that task across many steps, holding the context of a design through hours-long workflows. UST aims to catch design flaws earlier, speed up chip validation, and unify hardware and software in a single system.
The clearest example is a UST platform called iDEC, which its engineers use to validate hardware and silicon before production. Validation is the work of proving a chip actually behaves the way its designers intended, and it's arduous: engineers write test scripts by hand, run them, read the results, and repeat the cycle many times. UST reports that iDEC's closed-loop pipeline-which reads hardware designs, generates and runs regression tests, and compares live equipment data against a digital twin to flag issues early-already cuts validation cycle times by 50 to 70%, condensing standard four-day turnarounds into 48 hours.
UST is now integrating Claude into that pipeline as its reasoning layer. Claude Code reads chip pinouts and hardware schematics directly, then writes and runs regression tests-the checks that confirm a design change didn't cause an unintended downstream effect-which engineers previously scripted by hand. Claude also compares live data from real equipment against its digital twin-the software model of how the hardware is supposed to behave-and flags firmware regressions and signal-integrity faults. UST's goal is to make the pipeline even faster, with less manual scripting, earlier fault detection, and no new tools for engineers to learn.
UST CEO Krishna Sudheendra stated that the alliance with Anthropic reflects UST's commitment to helping clients navigate the AI landscape with confidence and achieve meaningful business outcomes. By combining Claude's capabilities with UST's engineering, industry knowledge, and delivery expertise, the companies are bringing industry-specific platforms and solutions to market that improve productivity, accelerate business outcomes, and help clients operationalize AI-led decisions in a safe and secure environment.
Anthropic's Chief Commercial Officer Paul Smith noted that UST helps the world's banks, telecoms, and manufacturers adopt new technology. UST is proving Claude inside its own engineering first and training 20,000 of its own people on it before bringing it into the systems it builds and runs for clients.
How UST Is Deploying Claude in Healthcare, Telecom, and Banking Systems
UST is bringing Claude into three other platforms it operates for clients:
In healthcare, insurers and providers use UST CarePath to run member services, care management, and claims. Claude connects CarePath directly to its underlying claims and care systems, turning scattered health data into clear next steps for care teams. Every recommended action routes to a person for approval before reaching a member, and it stays within the data controls healthcare requires.
In telecom, UST IntelliOps runs network operations. To keep networks running, teams work through alerts that spot problems and outages, but this is a time-consuming process. Claude now helps operators spot service issues, predict failures in the radio access network (the towers and antennas connecting phones to the network), and shorten outages through response workflows that a person still approves. For teams monitoring alerts, that means less time separating real problems from noise.
In banking, most mid-sized institutions still run on core systems old enough that the ledger updates once a night rather than in real time. Banks often license these systems rather than own them, so every new product or integration means months of waiting for a vendor to make the change. UST FinX helps banks modernize progressively by solving immediate business and operational challenges while reducing dependency on disruptive, high-risk transformation programs. FinX will use Claude to embed AI agents directly into bank workflows and processes, supporting both operations teams and customers through intelligent case handling, servicing automation, knowledge retrieval, workflow assistance, and decision support.
A Partnership in AI Adoption and Training
UST is committing to train 20,000 of its associates on Claude worldwide, including engineers, architects, consultants, industry specialists, and forward-deployed engineers who sit alongside client teams. UST is also building specialized teams to deploy Claude. Anthropic is supporting UST in this effort with enablement, technical guidance, and certification through the Claude Partner Network. This partnership also makes UST a Global Premier Partner in the Claude Partner Network.
Pairing Reliability, Safety, and Governance
The industries UST serves are extremely high-stakes. Human approval steps and audit controls are important forms of governance that make it possible for systems to run in production. The reliability and safety Anthropic has prioritized in building Claude, paired with UST's experience in governance and regulated delivery, enables this work to move out of a pilot and into the systems that run a business.