Research Intern

About the Role

As an Applied Research Intern at Sierra, you will join the Applied Research team to work on a research project under the guidance of a mentor. You will identify a problem relevant to the real-world performance of AI agents, conduct rigorous experiments, and potentially write up your findings for a top academic venue.

Key Responsibilities

• Own a research project from conception to submission, including framing questions, running experiments, and writing a paper.
• Advance agent evaluation by extending τ-bench to new domains, modalities, or long-horizon tasks, or by developing new methods for measuring reliability and consistency.
• Study agent reasoning, tool usage, and policy adherence in multi-turn conversations, and identify failure points.
• Explore post-training methods such as supervised fine-tuning and reinforcement learning to enhance agent behavior, and rigorously measure improvements.
• Utilize large-scale, proprietary datasets of text and audio interactions to identify research questions grounded in real customer issues.
• Collaborate with researchers and engineers across Sierra, and translate findings into production improvements where applicable.

Requirements & Tech Stack

• Currently pursuing a PhD in computer science, machine learning, or a related field.
• A research track record, evidenced by first-author publications or submissions at venues such as NeurIPS, ICML, ICLR, ACL, EMNLP, Interspeech, or ICASSP.
• Strong experimental skills, including designing clean experiments, building careful baselines, and interpreting results accurately.
• Fluency in Python and modern ML tooling, including PyTorch, and LLM inference and training stacks.
• High agency and a builder's mindset, with the ability to work across the stack to execute experiments.
• Clear and direct communication skills, both written and in research presentations.

Even Better

• Research experience with LLM agents, tool use, or multi-turn dialogue.
• Experience designing benchmarks or evaluations, particularly those linked to real-world outcomes.
• Experience with supervised fine-tuning, reinforcement learning, or long-horizon optimization.
• Experience with speech, voice, or multimodal models.
• Open-source contributions to ML research code or benchmarks.

Intern Qualifications

• The internship is open to students currently enrolled in a PhD program.
• Duration and start dates are flexible.
• Interns will be embedded with the Applied Research team and work under the guidance of a mentor.
• The internship is based in our San Francisco office. In-person attendance is required Monday through Thursday.