Research Engineer focused on adapting and optimizing advanced language and vision models for efficient inference on Cerebras' wafer-scale AI architecture. The role involves implementing, validating, and optimizing models for low-latency, high-throughput inference, with a focus on techniques like speculative decoding, pruning, compression, and sparsity.

Kernel Engineer role focused on developing and optimizing high-performance software for Cerebras' AI chip, specifically implementing and scaling deep learning operations and building parallel algorithms for training and inference. The role involves low-level programming, performance tuning, and interaction with hardware architects to maximize compute utilization and accelerate AI innovation.