Together AI logo for Research Intern, Frontier Speculative Decoding (Winter 2027) job on DevsUnite

Research Intern, Frontier Speculative Decoding (Winter 2027)

Together AI

Actively HiringInternship
Apply
Location

San Francisco, CA

Experience

0-2 years

Role

Research Intern, Frontier Speculative Decoding (Winter 2027)

Salary

Competitive

Share this job

Know someone who'd be a great fit?

Job Description

Role Overview

The Speculative Decoding team at Together AI builds draft models and algorithms that enable large language models to generate faster without changing outputs. As a research intern, you will dive into open questions in speculative decoding: new drafter architectures, verification schemes, knowledge distillation, and training recipes that align draft and target models. You will train and fine-tune draft models at scale on dedicated GPU clusters and test ideas against realistic serving workloads, including long-context and agentic traffic.

Responsibilities

  • Research and prototype new efficient speculative decoding algorithms and draft model architectures
  • Train, fine-tune and distill draft models, and design rigorous experiments to validate hypotheses
  • Implement optimized versions for serving engines to show efficacy in real world workloads
  • Communicate plans, progress, and results of projects to the broader team
  • Share findings through technical write-ups and blog posts, and help bring successful ideas into production

Requirements

  • Currently pursuing a Master's or Ph.D. degree in Computer Science, Electrical Engineering, or a related field
  • Strong knowledge of Machine Learning and Deep Learning fundamentals
  • Hands-on experience training or fine-tuning large models with distributed training frameworks (e.g., Megatron-LM, FSDP)
  • Familiarity with modern model architectures, such as Mixture-of-Experts and hybrid attention

Required Skills

PythonPyTorchMegatron-LMFSDPvLLM

Ready to apply for Research Intern, Frontier Speculative Decoding (Winter 2027)?

See how you stack up at Together AI before you apply, or go straight to the application.

Apply Anyway