Full Stack LLM Engineer - Training

Cerebras Systems(2 days ago)

Sunnyvale, CA, Toronto, CanadaOnsiteFull TimeSeniorEngineering

About this role

Cerebras Systems is seeking an engineer to join their SOTA Training Platform team, focusing on bringing up state-of-the-art AI models on Cerebras CSX systems. The role involves working across the software stack to optimize performance and scalability, contributing to groundbreaking advancements in AI hardware and software.

View Original Listing

Required Skills

Python
C++
LLVM
MLIR
Performance Profiling
Deep Learning
Compiler Optimization
TensorFlow
PyTorch
Model Internals

About Cerebras Systems

cerebras.ai

Cerebras builds purpose‑built AI compute systems centered on its wafer‑scale processors to accelerate training and inference of large neural networks. Their integrated hardware‑and‑software platform delivers high throughput, low latency, and very large on‑chip memory/interconnect to shorten time‑to‑train for demanding AI workloads. Cerebras targets research labs and enterprises that need to scale experiments and deploy large models more quickly, pairing systems, tooling, and support to simplify large‑model development.

View more jobs at Cerebras Systems →