Research Scientist Intern, Ranking & Recommendations (PhD)

Meta•Menlo Park, CA

1d•$7,650 - $12,667

About The Position

Meta is seeking Research Interns to join its Post-Training & Research team within Facebook Discovery, which builds systems that determine which posts, photos, videos, and Reels people see on Facebook. The team works on ranking and recommendation, content understanding, and post-training AI models for agentic experiences on Facebook. Interns develop open problems into research questions and work with research scientists and engineers on work that may affect billions of people. The team studies generative recommendation models trained on sequences of user activity, ranking models that reason over a person's history and candidate content, multimodal models for assessing the quality and originality of posts and videos, and models post-trained using supervised fine-tuning and reinforcement learning from user feedback. Models are deployed with product and infrastructure teams, and results are published to advance the state of the art. Internships last 12 to 24 weeks, with various start dates throughout the year.

Requirements

Currently have or be in the process of obtaining a Ph.D. in Computer Science, Artificial Intelligence, Machine Learning, or a relevant technical field.
Obtain work authorization in the country of employment by the time of hire and maintain work authorization during employment.
Experience with Python and deep learning frameworks such as PyTorch.
Familiarity with modern AI/ML modeling techniques, such as Transformers, deep learning for recommendation, LLMs and LLM reasoning, multimodal models, mixture-of-experts, diffusion models, RAG, AI agents, and post-training with SFT and reinforcement learning.
Experience building systems based on machine-learning and/or deep-learning methods.

Nice To Haves

Intent to return to the degree program after completing the internship.
A track record of significant results demonstrated through grants, fellowships, patents, or publications at leading workshops or conferences such as NeurIPS, ICLR, ICML, KDD, SIGIR, WSDM, RecSys, CIKM, ACL, EMNLP, CoLM, CVPR, ICCV, ECCV, or similar venues.
Research experience in one or more of: LLM reasoning; post-training (SFT, reinforcement learning, reward modeling); AI agents; generative recommendation; diffusion models; ranking and recommendation systems; sequence and user modeling; computer vision; or multimodal content understanding.
Experience training large-scale deep learning models, such as LLMs or recommendation models.
Experience with C++ or optimizing LLM training and inference.
Demonstrated software engineering experience through an internship, work experience, coding competitions, or widely used open-source contributions, such as GitHub projects.
Proficiency using AI coding agents for rapid development and independent problem solving.
Experience working and communicating cross-functionally in a team environment.

Responsibilities

Define and lead a machine-learning research project in ranking and recommendation, with clear milestones throughout the internship.
Design, train, and post-train deep learning models, from recommendation models to LLMs and AI agents, using Facebook-scale data.
Evaluate models using offline metrics and online experiments, and analyze results to determine subsequent iterations.
Publish research results and contribute research that can be applied to Meta product development.
Collaborate with researchers and cross-functional partners, communicating research plans, progress, and results.