Systems/Rack Architect

Normal Computing | Build with Us

Normal is an applied AI company solving the hardest problems in AI and silicon. We build foundational hardware and software for the semiconductor industry, critical AI infrastructure, and the broader systems that power our world, in partnership with the world's most advanced institutions. We work as one team across New York City, Silicon Valley (Mountain View), and London.

The Role

Hybrid: 3 days a week in the office, in Mountain View, New York or London.

Normal’s hardware does not look like just another GPU server, so the system around it cannot be copied from one. As our Hardware System Design Architect, you will lead the system architecture design of Normal’s AI compute hardware end-to-end, from initial concept to specification, design reviews, integration, and deployment. You will own system level simulation and modeling and you will use this to guide system-level decisions spanning boards, chassis, interconnects, and rack-scale system design, collaborating closely across hardware, software, and other engineering teams to ensure system designs meet performance and workload requirements.

Working closely with our silicon performance modeling team, software, and lab teams, you will define system requirements and specifications, evaluate architectural options, and drive key technical decisions as our hardware platforms evolve. You’ll evaluate partner and vendor designs against our requirements, identify technical gaps and schedule risks early, and lead system-level tradeoffs. You’ll play a central role in bringing these pieces together into cohesive, high-performance AI compute systems and driving the overall system from architecture through deployment.

What You Will Own

• System Architecture and Specifications: Own path-finding and architecture for Normal’s compute system from conception to rack scale deployment. Write the system, board, interface, and rack specifications that internal teams and partners build against.
• Workload-to-System Translation: Translate workload and performance requirements into system specifications: accelerators per node, their topology, memory capacity, and bandwidth per node as well as the performance of scale-up and scale-out fabrics. Own system level performance, power, and total cost of ownership and compare candidate architectures with the needs of customers and the offerings of competitors. Partner with internal teams to model and simulate these performance tradeoffs.
• Simulation and Modeling: Own the simulations and hardware models behind every platform decision. Run what-if scenarios to identify the best designs and validate deployments in simulation. Consider diverse toolsets from CFD of airflow, coolant loops, rack power distribution, as well as network and workload simulation of scale-out traffic, reliability models, and serviceability.
• Boards, Chassis, Power and Physical Design: Drive board and chassis-level architecture from major component decisions to power delivery, and thermal and mechanical design. Define the physical, electrical, and thermal envelopes of our design.
• Interconnect and Networking: Define scale-up and scale-out fabrics from PCIe/CXL interfaces, to card-to-card links, NICs, switches, and optical cabling. Link power, signal integrity, and networking performance requirements holistically with the physical design.
• Partners and Suppliers: Review ODM, JDM and supplier designs against our requirements, surface gaps, and schedule risk early enough to act on and drive issues to resolution.
• Bring-up and Qualification: Define and own the bring-up, integration, validation, and qualification strategy for each platform including roof-of-concept builds that de-risk new architecture before volume production.

What Makes You a Great Fit

• Significant experience architecting complex compute hardware systems, spanning server or machine-level designs through rack-scale platforms
• Experience taking hardware systems from early requirements and architecture through specification, integration, bring-up, validation, and production deployment
• Degree in Electrical Engineering, Computer Engineering, or a related field
• Excellent written and interpersonal communication skills

Bonus Points

• Successfully architected multiple shipping commercial systems
• Built new performance, electrical, cooling models from scratch
• Substantial interactions with partners, including finding new partners and suppliers

Equal Employment Opportunity Statement

Normal Computing is an Equal Opportunity Employer. We celebrate diversity and are committed to creating an inclusive environment for all employees. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, veteran status, or any other legally protected status.

Accessibility Accommodations

Normal Computing is committed to providing reasonable accommodations to individuals with disabilities. If you need assistance or an accommodation due to a disability, please let us know at [email protected].

Privacy Notice

By submitting your application, you agree that Normal Computing may collect, use, and store your personal information for employment-related purposes in accordance with our Applicant Privacy Notice.

Compensation: $200K - $400K