Research Scientist Intern, User Modeling and Reinforcement Learning (PhD)

Meta; Bellevue, Menlo Park (WA, CA); On-site; USD 55,000 - 83,000; Full time

Job summary

Meta in Bellevue, WA seeks Research Scientist Interns for the Core Ads Growth group. You will own a focused research project and work closely with researchers and engineers to bring ideas into production-scale experiments at Meta's ads systems.

Internships are 12–16 or 24 weeks with various start dates, focusing on representation learning, user intent evolution, and ad-system improvements. Opportunities to publish and contribute to product development are available.

Qualifications

• Currently pursuing or in process of obtaining a Ph.D. in Computer Science, Machine Learning, Artificial Intelligence, Reinforcement Learning, Statistics, Applied Mathematics, or related field
• Must have work authorization in the United States at the time of hire and maintain it during employment
• Experience with deep learning frameworks such as PyTorch or TensorFlow
• Experience with Python, C++, or another relevant programming language
• Experience conducting research and building systems based on ML or DL methods
• Intent to return to a degree program after completion of the internship/co-op
• Experience in self-supervised learning, transformers, temporal sequence modeling, representation learning, or recommender systems
• Experience in reinforcement learning, causal inference, counterfactual evaluation, or learning from delayed and censored outcomes
• Proven track record of significant research results (grants, fellowships, patents, first-authored publications)
• Experience analyzing high-volume, high-dimensional data and designing controlled ML experiments
• Software engineering experience through internships, research implementations, or open-source contributions
• Self-directed quantitative problem-solving and cross-disciplinary teamwork

Responsibilities

• Own a focused research project with mentorship from researchers and engineers
• Translate new ideas into rigorous experiments and production-scale ideas in Meta's ads systems
• Contribute to scientific publications and product development

Skills

Deep learning research

Experiment design

Data analysis

Cross-disciplinary collaboration

Education

Ph.D. in CS/ML/AI or related field

Tools

PyTorch

TensorFlow

Python

C++

Job description

Meta is seeking Research Scientist Interns to join Core Ads Growth (CAG), where we build AI that improves ad relevance and helps advertisers succeed. Our research focuses on how user intent evolves across interactions and how ads systems can make better decisions when outcomes unfold over days, weeks, or longer.

We are advancing representation learning to better understand how people's interests and intent evolve over time. You will help bring new research ideas into production systems at Meta's scale, shaping more relevant experiences and creating lasting value for people and businesses.

As an intern, you will own a focused research project with close mentorship from researchers and engineers. The work connects new ideas to rigorous experiments and applications in Meta's ads systems, with opportunities to contribute to scientific publications and product development.

Our internships are twelve (12) to sixteen (16), or twenty-four (24) weeks long, with various start dates throughout the year.

• Currently has or is in the process of obtaining a Ph.D. degree in Computer Science, Machine Learning, Artificial Intelligence, Reinforcement Learning, Statistics, Applied Mathematics, or a related technical field
• Must obtain work authorization in the country of employment at the time of hire and maintain ongoing work authorization during employment
• Experience with deep learning frameworks such as PyTorch or TensorFlow
• Experience with Python, C++, or another relevant programming language
• Experience conducting research and building systems based on machine learning or deep learning methods
• Intent to return to a degree program after completion of the internship/co‑op
• Experience in one or more of self‑supervised learning, transformers, temporal sequence modeling, representation learning, or recommender systems
• Experience in reinforcement learning, causal inference, counterfactual evaluation, or learning from delayed and censored outcomes
• Proven track record of achieving significant research results as demonstrated by grants, fellowships, patents, as well as first‑authored publications at leading workshops or conferences such as NeurIPS, ICML, ICLR, KDD, RecSys, and similar conferences or workshops
• Experience manipulating and analyzing complex, high‑volume, high‑dimensional data and designing controlled machine learning experiments
• Demonstrated software engineering experience through internships, research implementations, or open‑source contributions
• Self‑directed quantitative problem‑solving and experience working and communicating across disciplines in a team environment