Research Intern, Frontier Agents (Winter 2027)

Location: San Francisco Department: Research

About the Role

The Agents team investigates how to build, align, and scale frontier AI systems that can tackle complex, multi-step tasks and workflows across text and speech, with a particular focus on agentic and scientific domains. Our work sits at the intersection of agent capabilities, human-computer interaction, and infrastructure—from designing post-training methods for agentic behavior to developing evaluation frameworks for open-ended tasks where traditional metrics fall short.

As a research intern, you will work on problems at the frontier of agentic AI, where challenges in alignment, reliability, and scalability are deeply intertwined. Projects may involve developing new training recipes for self-learning and long-horizon reasoning, curating datasets for non-deterministic scientific and agentic tasks, studying failure modes in agentic behavior, or building infrastructure that enables agent operations at scale. You'll operate in a space where algorithmic innovation, dataset and interaction design, and systems work come together to push the boundaries of what AI agents can reliably accomplish.

Responsibilities

Research and implement novel techniques in one or more of our focus areas

Design and conduct rigorous experiments to validate hypotheses

Document findings in scientific publications and blog posts

Communicate the plans, progress, and results of projects to the broader team

Requirements

Currently pursuing a Ph.D. degree in Computer Science, Electrical Engineering, Information Science, or a related field

Publications at leading ML, NLP, or speech conferences or journals (such as NeurIPS, ICML, ICLR, *ACL, EMNLP, Interspeech)

Strong knowledge of Machine Learning and Deep Learning fundamentals

Experience with deep learning frameworks (PyTorch, JAX, etc.)

Understanding of how LLMs work

Strong programming skills in Python

Familiarity with Transformer architectures and recent developments in foundation models

Example Research Directions

Training, developing, and evaluating frontier models, especially in the domain of agentic tasks and workflows

Designing and curating datasets for frontier agents alignment and post-training

Studying failure modes and developing safety paradigms for agentic behavior

Research new recipes (for RL or test time scaling) for self-learning and long-context tasks completion

Building agents that act on spoken input to carry out complex, multi-step tasks

Developing ML infrastructure that can power agent operations at scale

About Together AI

Together AI, the AI Native Cloud, is purpose-built for AI engineers. AI application developers get high-performance inference that scales reliably, fine-tuning and reinforcement learning for creating frontier-level specialized models, and pre-training at massive scale for fully custom intelligence, all around a marketplace of leading open models that teams can run, adapt, and own. Trusted by Cursor, Decagon, ElevenLabs, Salesforce, and Zoom, Together serves 400+ trillion tokens a month.

Internship Program Details

Our internship program runs 12 to 14 weeks, giving you the opportunity to work alongside industry-leading engineers and researchers across multiple teams. This cohort's internship dates span January 4th to April 9th.

Compensation

We offer competitive compensation, housing stipends, and other competitive benefits. The estimated US hourly rate for this role is $58 - $70 an hour. Our hourly rates are determined by location, level and role. Individual compensation will be determined by experience, skills, and job-related knowledge.

Equal Opportunity

Together AI is an Equal Opportunity Employer and is proud to offer equal employment opportunity to everyone regardless of race, color, ancestry, religion, sex, national origin, sexual orientation, age, citizenship, marital status, disability, gender identity, veteran status, and more.

For more information about our privacy policy, please visit: https://www.together.ai/privacy