San Francisco, CA
0-2 years
Research Intern, Frontier Speculative Decoding (Summer 2027)
Competitive
The Speculative Decoding team at Together AI develops draft models and algorithms to accelerate large language model generation without altering outputs. As a research intern, you will explore speculative decoding research questions, including new drafter architectures, verification schemes, and adaptive drafters for evolving workloads. You will train and evaluate draft models on GPU clusters, testing against realistic serving scenarios like long-context and agentic traffic.
See how you stack up at Together AI before you apply, or go straight to the application.