Applied AI/ML Scientist
Core
Develop and customize large language models and deep learning models to solve specific customer problems using Cerebras Wafer-Scale Engine hardware.
Role type
Applied AI/ML Scientist (Customer-Facing)
Builds
Custom SOTA models, agentic systems, and training pipelines for enterprise clients
Domain
AI/ML, Large Language Models, Custom Hardware (Wafer-Scale Engine)
Deliverable
production ML models
Required skills
Large model training (1B+ parameters), fine-tuning (SFT, RLHF, DPO), data curation, distributed training frameworks, Python, PyTorch, model architecture design, loss dynamics analysis, hardware scaling
Preferred skills
Experience with MoEs, multimodal models, agentic system components, tool-use capabilities, long-context reasoning
Technologies
Cerebras Wafer-Scale Engine (WSE), Python, PyTorch
Responsibilities
Collaborate with stakeholders to scope AI projects and define success metrics; Architect and execute end-to-end training recipes for custom models; Design adaptation strategies including continuous pre-training and alignment; Take ownership of training pipelines from data preprocessing to hyperparameter tuning; Scale training workloads across Cerebras clusters; Serve as AI/ML subject matter expert for technical deep-dives; Partner with internal teams to drive software stack improvements and distill customer projects into playbooks
Seniority
Senior, hands-on IC with customer leadership
