The opportunity
Cerebras Systems builds the world's largest AI chip, 56 times larger than GPUs. This architecture allows Cerebras to deliver industry-leading training and inference speeds; over 10 times faster than GPU-based hyperscale cloud inference services.
What you'll do
Customer Use Case Discovery & Project Scoping Collaborate with customer: stakeholders to identify the best approaches to their business problem with AI.
Contribute to the technical scoping of engagements, including feasibility: analysis, data quality/availability/readiness assessments, and the selection of optimal model architectures.
Define project milestones, success metrics, and rigorous evaluation: benchmarks to ensure the solution delivers measurable value to the customer’s business.
Custom SOTA Models and AI Systems Development Architect and: execute end-to-end training recipes for custom models, tailoring model architecture and training recipes to meet customer-specific performance and accuracy requirements.
Design and implement sophisticated adaptation strategies, including: continuous pre-training on private datasets, supervised fine-tuning (SFT), and post-training alignment via RLHF or DPO.
Take full ownership of the training pipeline, from high-performance data: preprocessing and tokenization to hyperparameter tuning and loss-curve analysis.
What they're looking for
- Navigate the nuances of model convergence on specialized hardware, performing: deep-dive analysis into loss dynamics and gradient stability.
- Scale training workloads across Cerebras clusters, ensuring: efficient utilization of the hardware for multi-billion parameter models.
- Build and optimize the core components of agentic systems, focusing on: tool-use capabilities, long-context reasoning, and multi-step planning.
- Technical Customer Leadership Serve as an AI/ML subject matter expert: during technical deep-dives, translating customer requirements into precise training recipes.
