Cerebras Systems

Build the ML API layer for Cerebras’s high-performance, disaggregated AI inference platform. This senior individual-contributor role spans model serving, API design, distributed systems, performance optimization, and production reliability.

Lead the technical direction and evolution of Cerebras’s observability platform at massive AI-computing scale. Build telemetry, instrumentation, alerting, and developer tooling across hardware and software systems.

Build and optimize Cerebras’s production GPU inference stack across APIs, model-serving runtimes, AMD software, distributed systems, and rack-scale infrastructure. This hands-on Staff Software Engineer role focuses on reliability, numerical correctness, and exceptional inference performance.

Cerebras is seeking a hands-on Senior SDET to build sophisticated automation and lead release integration testing for complex AI inference systems. The role spans frameworks, runtimes, compilers, distributed systems, infrastructure, and hardware while improving production release reliability.

Own the full model bring-up process from reference implementation to efficient execution on Cerebras AI accelerator hardware. You’ll combine ML engineering, compiler development, kernel optimization, and performance analysis to deliver fast, numerically correct inference.

Build the software platforms and operational tools that monitor, manage, and scale Cerebras’s growing fleet of AI clusters. This senior role spans backend services, infrastructure integrations, operator-facing applications, and reliability at large scale.

Join Cerebras’s Core ML team to map emerging machine learning algorithms onto its wafer-scale architecture and determine how they perform across models, workloads, and hardware scales. The role combines performance modeling, benchmarking, prototyping, and hardware-software co-design to shape future AI systems.

Build high-performance ML runtimes and kernels that turn Core ML research ideas into efficient training and inference capabilities on Cerebras systems. The role spans frameworks, compilers, distributed execution, performance optimization, and low-level kernel development.