Intern: Frontier Speculative Decoding Research Engineer

Together Computer Inc; San Francisco (CA); On-site; USD 80,000 - 96,000; Full time

Benefits offered by this job

Housing stipend

Competitive compensation

Paid internship

Job summary

Together AI in San Francisco is seeking a research intern for the Speculative Decoding team to explore efficient draft models and verification schemes. You will train and fine-tune draft models at scale on dedicated GPU clusters, testing ideas against production-style workloads.

Ideal candidates are pursuing a Master’s or PhD in CS/EE with strong ML/DL fundamentals, hands-on experience with distributed training (Megatron-LM, FSDP), and familiarity with Mixture-of-Experts and hybrid attention.

Qualifications

• Pursuing a Master’s or PhD in CS/EE, or related field.
• Strong ML/DL fundamentals.
• Hands-on experience training or fine-tuning large models with distributed frameworks.
• Familiarity with Mixture-of-Experts and hybrid attention.

Responsibilities

• Research speculative decoding architectures and verification schemes.
• Train, fine-tune draft models at scale on GPU clusters.
• Evaluate ideas against realistic serving workloads.
• Communicate progress and results to the team.

Skills

ML fundamentals

Distributed training

MoE architectures

Hybrid attention

Education

Master's or PhD in CS/EE

Tools

Megatron-LM

FSDP

Job description

Together AI in San Francisco is seeking a research intern for the Speculative Decoding team to explore efficient draft models and verification schemes. You will train and fine-tune draft models at scale on dedicated GPU clusters, testing ideas against production-style workloads.

Ideal candidates are pursuing a Master’s or PhD in CS/EE with strong ML/DL fundamentals, hands-on experience with distributed training (Megatron-LM, FSDP), and familiarity with Mixture-of-Experts and hybrid attention.