Senior Research Engineer - Inference ML
Core
Adapt advanced language and vision models to run efficiently on Cerebras architecture, focusing on speculative decoding, pruning, compression, and sparse attention to deliver low-latency, high-throughput inference.
Role type
Senior Research Engineer (Inference ML)
Builds
Optimized inference workloads for NLP and computer vision models on Cerebras hardware
Domain
AI Hardware / Machine Learning / Inference Optimization
Deliverable
production ML models
Required skills
Transformer architecture design, speculative decoding, model pruning and compression, sparse attention, Python, C++, PyTorch, vLLM, SGLang, performance profiling, hyperparameter tuning
Preferred skills
Triton/CUDA, mixture-of-experts, multimodal learning, AI agents, large-scale model deployment, quantization
Technologies
Cerebras Wafer-Scale Engine, PyTorch, Transformers, vLLM, SGLang, Triton, CUDA
Responsibilities
Design and optimize transformer architectures for NLP and CV; research and prototype novel inference algorithms; train models and perform hyperparameter sweeps; bring up new models and validate functional correctness; profile and optimize model code; develop diagnostic tooling for performance bottlenecks; collaborate across software, hardware, and product teams
Seniority
Senior, hands-on IC
