Senior GPU Inference Systems Engineer

Cerebras Systems; California (MO); On-site; USD 180,000 - 260,000; Full time

Job summary

Cerebras Systems is hiring a Software Engineer to productionize and optimize the GPU serving stack across custom APIs, vLLM, ROCm, and rack-scale GPU infrastructure.

You will write production code, establish operational practices, and drive improvements in time to first token, throughput, tail latency, and capacity efficiency. This hands-on role spans application, runtime, distributed systems, and hardware layers.

Qualifications

• 8+ years of software engineering experience in production systems.
• Experience with production inference systems for large language models or similar GPU workloads.
• Strong programming ability in C++ and Python with multithreading and memory management.
• Hands-on experience with a high-performance model-serving framework (vLLM, SGLang, TensorRT-LLM, Triton or equivalent).
• Strong understanding of GPU execution, profiling, and asynchronous operations.
• Experience debugging distributed systems across multiple layers.
• Experience with Linux, containers, Kubernetes, CI/CD, and latency-sensitive services.
• Ability to design benchmarks and interpret noisy results for production improvements.
• Strong communication and technical leadership skills.
• Bachelor’s degree in Computer Science/Engineering or equivalent.

Responsibilities

• Productionize the GPU inference stack across API services, vLLM, ROCm, and hardware stack.
• Own GPU operational readiness: deployment, upgrades, health checks, capacity management.
• Drive reliability in production: define SLIs/OKRs, incident response, automated recovery.
• Improve inference performance: reduce time to first token, increase throughput, lower tail latency.
• Tune model-serving behavior: scheduling, continuous batching, KV-cache, parallelism, quantization.
• Debug across system layers: diagnose failures across apps, libraries, drivers and hardware.
• Ensure numerical correctness with validation and regression infrastructure.
• Build benchmarks, profiling automation, and release gates.

Skills

C++

Python

Distributed systems

GPU/ML serving

Linux

Education

Bachelor's degree

Tools

vLLM

SGLang

TensorRT-LLM

Triton

ROCm

Job description

Cerebras Systems is hiring a Software Engineer to productionize and optimize the GPU serving stack across custom APIs, vLLM, ROCm, and rack-scale GPU infrastructure.

You will write production code, establish operational practices, and drive improvements in time to first token, throughput, tail latency, and capacity efficiency. This hands-on role spans application, runtime, distributed systems, and hardware layers.