Research Intern, Model Shaping (Winter 2027)
About this role
Investigating advanced post-training methods and efficient neural network training forms the daily focus for this Winter 2027 internship. The Model Shaping team builds services and conducts research that enable machine learning developers to adapt open foundation models for downstream applications using domain-specific data. Members of this group also create novel techniques for model evaluation and efficient training by drawing on ideas from machine learning, natural language processing, and ML systems.
Created to support ongoing research initiatives, the position gives candidates the chance to work alongside industry-leading engineers and researchers during a 12 to 14 week cohort. Successful candidates generally combine strong academic foundations in computer science or related fields with practical experience using deep learning frameworks and foundation models.
What you will do
- Research and implement novel techniques in supervised learning, preference optimization, or reinforcement learning.
- Design and conduct rigorous experiments to validate research hypotheses.
- Document experimental findings in scientific publications and blog posts.
- Integrate successful research results directly into Together products.
What you need
- Must be currently pursuing a Bachelor's, Master's, or Ph.D. degree in Computer Science, Electrical Engineering, or a related field.
- Demonstrate strong knowledge of machine learning and deep learning fundamentals.
- Bring practical experience with deep learning frameworks such as PyTorch or JAX.
- Maintain familiarity with the Transformer architecture and recent developments in foundation models.
Nice to have
- Prior research experience with training foundation models or efficient machine learning
- Publications at leading ML and NLP conferences including NeurIPS, ICML, ICLR, ACL, or EMNLP
- Understanding of model optimization techniques and hardware acceleration approaches
- Contributions to open-source machine learning projects
Benefits
- Competitive compensation
- Housing stipends
- Other competitive benefits
Skills and tech stack
Questions candidates ask
What are the dates and locations for the internship?
The internship runs from January 4th to April 9th. The role is based either in the San Francisco or Amsterdam office.
What is the compensation for this role?
The estimated US hourly rate for this role is $58 to $70. Compensation is determined by location, level, role, experience, skills, and job-related knowledge.
How long is the internship program?
The internship program runs 12 to 14 weeks. It provides the opportunity to work alongside industry-leading engineers and researchers across multiple teams.
What does the Model Shaping team actually build?
The team builds services that allow machine learning developers to choose the best models for their tasks. They also build products for tailoring open foundation models to downstream applications using domain-specific data.