Product Engineer - Manufacturing Operations

About Anthropic

Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.

About the role

Anthropic is building its own servers, racks, and accelerator systems with contract manufacturers and ODM partners, and deploying them across our fleet of data centers. We are seeking experienced engineers to own how well that hardware gets built.

As a Product Engineer on the Manufacturing Operations team, you are the technical owner of a product at the factory. During development, you gather manufacturing, test, and yield concerns, get them fixed in the design before each build, and sign off that the product is ready to move to the next stage. Once a product is in mass production, you keep it healthy by delivering engineering changes with minimal disruption, qualifying new components and second-source manufacturers, and root-causing yield and field failures.

This is a hands-on role with up to 50% travel to partner factories in the US and abroad. You should be as comfortable debugging a test failure on the line as you are presenting a corrective action plan to leadership.

Representative projects

Take a new accelerator rack from early evaluation to mass production at an ODM: audit the assembly and test process, set yield targets, confirm test coverage, and approve the transition.

Scale manufacturing from one line to multiple and then to a second site or second manufacturer without yield or throughput regressions.

Run a cross-functional root cause on field failures traced to a manufacturing gap, drive corrective action with the ODM/CM, and close the loop with the design itself when required.

Lead an engineering change across manufacturers: BOM and PLM updates, cut-in planning, partner training, validation, and monitor field failures/RMA.

Bring manufacturing requirements and validation data into each development phase review so the engineering team has a single prioritized list of what to fix before the next build.

You may be a good fit if you

Have carried a hardware product from early builds through mass production at a contract manufacturer or ODM.

Have set up or improved assembly and test processes at a manufacturing partner and drove yield, quality, and schedule commitments.

Use structured manufacturing and root-cause methods (DFM, FMEA, SPC, 8D or similar) as a matter of habit, and are an expert at translating messy failure data into clear decisions.

Can write a product-specific test plan and judge whether test coverage is sufficient.

Communicate clearly with partner engineers and operations staff on the floor as well as with leadership in a readout.

Are willing to travel to partner sites up to 50% of the time.

Have a degree in Electrical, Mechanical, Industrial, or Materials Engineering, or equivalent practical experience.

Strong candidates may also

Have been the technical lead for a data center hardware program such as: server, storage, switch, rack, or accelerator.

Have stood up a new manufacturing line or site, or brought up a second CM for an existing product.

Have experience with liquid-cooled or high-power-density rack systems.