Job Specifications
Netflix is one of the world's leading entertainment services, with over 300 million paid memberships in over 190 countries enjoying TV series, films and games across a wide variety of genres and languages. Members can play, pause and resume watching as much as they want, anytime, anywhere, and can change their plans at any time.
Machine Learning Research at Netflix improves various aspects of our business, including personalization algorithms, member and title understanding, creative tooling, system optimization, and innovative tooling. Our research spans many areas of machine learning, including recommender systems, reinforcement learning, computer vision, natural language processing, optimization, causality, and operations research. Great applied research also requires robust machine learning infrastructure, another strong emphasis at Netflix.
Candidates will be evaluated to find the best fit in one of our organizations, including Content, Choosing & Conversation, Commerce or AI for Member Systems. You can find a detailed list of teams across these organizations to learn more. Applicants are encouraged to express their interest in one or multiple types of teams/ domain areas listed if your skills and qualifications are aligned.
We Are Looking For Individuals With The Following Qualifications
Currently enrolled student pursuing an advanced degree (PhD) in areas such as Computer Science, Machine Learning, Artificial Intelligence, Computer Engineering, Mathematics, Statistics, Data Science, Economics, Computational Biology, Chemistry, Physics, Cognitive Science or a related field
Domain expertise in one or more of the following areas:
Personalization & Recommender Systems: Using Transformers/LLMs for recommendations, collaborative filtering, content-based recommendation, hybrid systems, and conversational recommenders.
Natural Language Processing (NLP): Large Language Models (LLMs), fine-tuning, in-context learning, prompt engineering, alignment, evaluation, text generation, and embeddings.
Reinforcement Learning (RL): Offline and online RL, alignment and post-training, preference- and human-feedback-based learning, bandit algorithms.
Computer Vision (CV): Image and video understanding, generation, and representation learning.
Computer Graphics: 3D modeling and understanding, neural rendering, animation, and related areas.
Reliable ML: Robustness, explainability, interpretability.
Causal ML: Causal inference, causal discovery, double ML, policy learning, dynamic panel and dynamic choice modeling, matrix completion for counterfactuals.
Agentic AI: Developing and evaluating agentic systems that reason, plan, and act autonomously, including tool use, retrieval-augmented reasoning, memory and goal management, and feedback-driven learning.
Multimodal Data: Experience in large vision language models, modality fusion and alignment, multimodal retrieval. Experience handling and integrating text, image, video, audio, and other data sources.
Model Optimization and Efficiency: Training and inference efficiency, model benchmarking, optimization techniques.
ML Platform & Infrastructure: Designing and building scalable systems for model development, training, and deployment, managing large-scale data pipelines and distributed compute environments.
General ML Application Engineering: Implementing machine learning solutions across various domains, end-to-end ML pipelines, from experimentation to deployment.
Experience programming in at least one programming language (Python, Java, Scala, or C/C++)
Familiarity developing ML models using common frameworks (e.g., PyTorch, TensorFlow, Keras) and training on GPUs.
Familiarity with distributed training and inference paradigms and associated frameworks (eg. DDP, FSDP, HSDP, Deepspeed)
Familiarity with end-to-end machine learning pipelines (e.g. training or production deployment) and common challenges like explainability.
Curious, self-motivated, and excited about solving open-ended challenges at Netflix.
Great communication skills, both oral and written.
Nice To Have
Comfortable with distributed computing environments such as Spark or Presto.
Comfortable with software engineering best practices (e.g. version control, testing, code review, etc.).
For your application to be considered complete
You will be sent an Airtable form shortly after you submit your application on our careers site; your application will not be considered complete until you fill out and submit this form.
Include a Resume or CV with complete contact information (email, phone, mailing address) and a list of relevant coursework and publications (if applicable).You will be asked to include a short statement describing your research experiences and interests, and (optionally) their relevance to Netflix Research. For inspiration, have a look at the Netflix Research site.
Applications will be reviewed on a rolling basis and it’s in the applicant's best interest to apply early. The application window will rema