Compute Server Platform Architect
Cerebras · Sunnyvale, CA
About this role
## Compute Server Platform Architect ### About the Role As a **Compute / Server Platform Architect** on the **Cluster Architecture Team**, you will own the **server-side platform architecture** that enables **Cerebras CS3-based AI clusters** (training and inference) to deliver **predictable performance, scalability, and reliability**. Because our accelerators are **network-attached**, the **x86 server fleet** is a first-class part of the end-to-end system. It runs critical-path runtime functions (e.g., **orchestration, prompt caching, and IO/control services**) and must be co-designed with software for **token-level latency, throughput, and cost efficiency**. You will translate workload behavior into **CPU, memory, IO, PCIe, and host-networking** requirements, drive **platform evaluations** with vendors, and provide technical leadership through **qualification and production adoption**. ### Responsibilities - Own the architecture for all **server roles** in Cerebras clusters, including **server types, configurations, and lifecycle strategy**. - Define and maintain **server formulas** (counts/ratios per CS-3 count, cluster size, workload type), including **capacity planning and headroom policy**. - Specify platform configurations, including: - **CPU SKU/core strategy** and vendor roadmap (e.g., **AMD, Intel, ARM**) - **Memory topology** (channels, DIMM type, capacity) - **PCIe topology** and **lane budgeting** - **NIC selection/placement** - **Local NVMe policy** where applicable - Translate software/runtime flows into measurable hardware requirements (CPU utilization, memory bandwidth/latency, bursty IO patterns, queueing/concurrency limits) and communicate clear guardrails to software teams. - Develop **performance and scaling models**; validate with microbenchmarks and workload-level experiments; identify bottlenecks and drive cross-stack fixes. - Define the **OS, BIOS, firmware, and driver baseline** for each server type (other teams will follow these recommendations for fleet configuration). - Stay current on emerging server technologies (CPU generations, memory tech, **CXL**, NVMe evolutions, **SmartNIC/DPU** capabilities where relevant) and run proof-of-concept evaluations to determine adoption timing. - Lead technical vendor engagements (OEM/ODM and component vendors): influence roadmaps, request platform knobs, and drive joint debugging on performance/reliability issues. - Define **qualification and acceptance criteria** (performance, stability, operability) and partner with the Infrastructure Hardware TPM to execute qualification plans and land changes into production. - Support bring-up and rare deployment debugging in lab/staging; drive root-cause analysis for regressions across firmware, drivers, OS, and runtime behavior. ### Skills & Qualifications - **PhD** in Computer Science or Electrical/Computer Engineering and **8+ years** industry experience, **or** Master’s/Bachelor’s in CS or EE and **10+ years** industry experience. - **5+ years** in server platform architecture, systems performance engineering, or large-scale infrastructure design for **AI/ML, HPC**, or performance-sensitive distributed systems. - Deep understanding of **x86 server architecture** (CPU microarchitecture basics, cache hierarchies, **NUMA**, memory controllers/channels, bandwidth vs latency tradeoffs). - Strong **Linux** systems knowledge (profiling/performance analysis, scheduling and syscall overheads, memory management behavior, pr
Listing freshness
CronJobs last confirmed this listing 57m ago. If its source stops confirming the opening for seven days, this page is removed from active inventory.