Research Intern, Frontier Speculative Decoding (Summer 2027)

What you’d do

Research intern on the Speculative Decoding team, working on drafter architectures, verification schemes, and distillation recipes, training and fine-tuning draft models on GPU clusters and testing them against realistic serving workloads. Found on the employer's careers page on Oct 21, 2026. Apply through the link.

You’ll need

• Pursuing Master's or Ph.D. in CS, EE, or related field
• Strong ML and deep learning fundamentals
• Hands-on large model training or fine-tuning
• Distributed training frameworks (e.g., Megatron-LM, FSDP)