Research Intern, Frontier Speculative Decoding (Winter 2027)
Together AI · San Francisco
Infrastructureonsiteintern$121k–$146k
At a glance
Highlights
- Build draft models that make frontier LLMs generate faster without changing outputs.
- Train draft models at scale and test ideas against realistic serving workloads.
- Help bring successful speculative decoding ideas into production systems.
What they offer
- Housing stipends
- Competitive benefits
Ideal candidate
A Master's or PhD student obsessed with efficient inference who enjoys co-designing models and systems and turning research prototypes into real-world speedups.
Requirements
- Currently pursuing a Master's or Ph.D. in Computer Science, Electrical Engineering, or related field
- Strong knowledge of Machine Learning and Deep Learning fundamentals
- Hands-on experience training or fine-tuning large models with distributed training frameworks
- Familiarity with modern model architectures such as Mixture-of-Experts and hybrid attention
Lab profile
About Together AI
AI-native cloud for training and inference on large-scale GPU infrastructure for AI apps and agents.
What they build
- AI-native cloud platform
- GPU clusters for training and inference
- Platform for AI applications and agents
Known for
- Operating large-scale GPU fleets
- Automated infrastructure at massive scale
- Multi-site data center deployment
Key facts
- Over 1M developers
- Operates one of the world's largest GPU fleets
- Infrastructure managing tens of thousands of GPUs
Posted · Last seen