How to apply for Data Center Controls Engineer

About Anthropic

Anthropic is a frontier AI research and product company working on alignment, policy, and security. It posts specific opportunities it considers high impact, and it points candidates to resources about the risks of working at a frontier AI lab. This role sits inside that context: building infrastructure that runs AI systems at scale.

About the role

This is a datacenter controls engineering role on Anthropic's Datacenter Infrastructure team. You would design and run control systems for liquid cooling (CDUs/CDMs), HVAC, electrical power monitoring, and building automation across multiple global sites. The work directly supports high-density AI compute, so reliability of control nodes, networks, and protocol layers matters.

A typical day

A typical day likely involves reviewing controls designs and protocol integrations across sites, coordinating with mechanical, electrical, and network teams, and troubleshooting control node or server availability issues. The job description does not detail daily routines, so ask the hiring team how time is split between design, integration, and operations.

Who Anthropic is looking for

• Has hands-on experience with industrial automation and building automation systems in datacenter or critical-facility environments.
• Knows BACnet, Modbus TCP/IP, OPC-UA, MQTT, and SNMP, and has designed standardized frameworks around them.
• Has built or maintained multi-site control architectures, including network topology, redundancy, and high-availability design for control nodes and servers.
• Can work across disciplines (mechanical, electrical, network, facilities) and translate controls requirements into integrated solutions for high-density compute.
• Has worked with liquid cooling systems such as CDUs and CDMs, or is ready to go deep on them.

Tips for this application

• Lead with multi-site controls architecture work. Anthropic explicitly lists multi-site design as a responsibility, so name the sites, scale, and protocols you used.
• Give concrete protocol examples. List which of BACnet, Modbus TCP/IP, OPC-UA, MQTT, and SNMP you have implemented, and what you built with each.
• Show redundancy and high-availability design. Describe specific control node or server HA setups you have delivered, including failover behavior.
• Mention liquid cooling directly. If you have CDU or CDM experience, say so in the first lines of your resume. If not, describe your closest thermal or mechanical controls work.
• Address the frontier AI context. Anthropic links to concerns about working at a frontier AI lab, so a short, direct statement about why you want to build infrastructure for this kind of company is worth including.

What to cover in your cover letter

Name the specific control systems you have built and at what scale. Explain how you have handled multi-site architecture, redundancy, and high-availability requirements. Describe your experience with the listed protocols and any liquid cooling or high-density compute work. State plainly why you want to work on datacenter infrastructure at an AI research company, given Anthropic's own note about the tradeoffs of working at a frontier lab.

Research before applying

• Read Anthropic's mission statement and its published work on alignment, policy, and security.
• Read the 80,000 Hours career review on working at an AI lab, which Anthropic links to directly in the job description.
• Look into the power and cooling demands of high-density AI compute, since the role centers on liquid cooling and datacenter controls.
• Check Anthropic's other posted roles and public infrastructure statements to understand how the Datacenter Infrastructure team fits into the company.

Likely interview topics

Based on the job description, expect questions about:

• How you would design a multi-site controls architecture with redundancy and high availability across global datacenters.
• Tradeoffs between BACnet, Modbus TCP/IP, OPC-UA, MQTT, and SNMP for different datacenter subsystems.
• Controls strategy for liquid cooling systems such as CDUs and CDMs under high-density AI compute loads.
• How you integrate control systems with electrical power monitoring and building automation systems.
• How you think about working at a frontier AI company, given Anthropic's own published concerns about doing harm in this space.

Common mistakes to avoid

• Treating this as a generic controls engineering job and skipping the AI datacenter context. The role is tied to high-density AI compute and liquid cooling.
• Listing protocols without showing what you built with them. Anthropic asks for standardized frameworks, not familiarity.
• Ignoring the multi-site aspect. Single-site experience alone does not match the stated responsibilities.