Full Stack LLM Engineer

Job Title

Full Stack LLM Engineer

Role Summary

Join the Inference Core Model Bringup team to rapidly bring up state-of-the-art open-source and customer models on Cerebras CSX systems. Work across the software stack to translate model architectures, optimize compilers and runtimes, and tune for high-performance inference.

This role focuses on system-level debugging, performance optimization, and integrating models end-to-end for production deployment.

Experience Level

Mid-level — no specific years required; expects demonstrated experience with ML model bringup, compiler/toolchain work, and low-level optimization.

Responsibilities

Primary responsibilities include:

• Bring up ML models end-to-end on Cerebras CSX systems (open-source and customer models).
• Translate model architectures and lower graphs into target compiler IRs and runtimes.
• Implement compiler optimizations and runtime integrations for performance and correctness.
• Profile, debug, and resolve performance, numerical accuracy, and runtime issues spanning model code, compiler IR, runtime behavior, and hardware utilization.
• Prototype and propose tooling, API, or automation improvements to accelerate future bring-ups.

Requirements

Must-have technical skills and experience:

• Proficiency with Python model code and deep learning frameworks such as PyTorch or TensorFlow; familiarity with model internals (attention, MoE, diffusion).
• Proven experience in compiler development and optimizations, especially with LLVM and/or MLIR.
• Strong C/C++ skills and experience with low-level optimization and runtime integration.
• Strong debugging and performance-profiling skills across numerical accuracy, runtime behavior, and hardware utilization.
• Background in optimization techniques for hard combinatorial problems.
• Comfort navigating the full AI toolchain: model code, compiler IRs, profilers, and runtime traces.

Nice-to-have:

• Experience with large-model bringup, model parallelism, and production inference at scale.

Education Requirements

Bachelor’s, Master’s, or PhD in Computer Science, Engineering, or a related field.

About the Company

Company: Cerebras

Headquarters: Sunnyvale, CA, USA

Developer of wafer-scale AI accelerators, Cerebras designs the Wafer Scale Engine (WSE)—one of the world’s largest AI chips—to deliver high-speed training and inference solutions for model labs, enterprises, and AI-native startups.