Research Engineer, RealTime AI, MSL PAR

Description

We are seeking research engineers to join the Product and Applied Research (PAR) Real Time AI group within Meta Superintelligence Labs (MSL). As a member of the PAR RealTime AI group, you will develop Large Language Models and foundational technology to build Meta’s AI Characters products. This is a high-impact Individual Contributor (IC) role focused on LLM model post-training, developing RAG/tool calling solutions, and partnering closely with cross functional teams to build the real-world experiences. We work directly with the product team and bring our state-of-the-art technology across FoA, and the entire Meta creator and developer ecosystem. If you’re excited about building the future of immersive real-time AI characters and passionate about making a tangible impact on billions of users, we invite you to join us.

Responsibilities

Collaborate with cross-functional teams to develop Meta’s AI Characters products Lead the development of new algorithms and systems for LLM post-training, evaluation and efficiency Support creative data sourcing, high-quality post-training data curation, and scale and optimize data pipelines for large language models (LLMs) Develop and integrate models,orchestrations and RAGs in production Analyze and interpret experimental results, iterate on model architectures, and drive continuous improvement Lead complex technical projects end-to-end

Qualifications

Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience 2+ years of industry experience in LLM/NLP, audio, or related AI/ML models Experience as a formal technical lead, leading major technical initiatives with cross functional partners to impact, and/or influencing strategy across multiple teams Skilled in model training, data, or inference & efficiency for LLMs Experience building products/systems based on machine learning, reinforcement learning and/or deep learning methods Programming experience in Python and hands-on experience with frameworks like PyTorch Experience with Reinforcement Learning from Human Feedback, reward modeling, or other LLM post-training techniques Experience manipulating and analyzing complex, high-volume, high-dimensionality data from varying sources Proven track record of achieving significant results as demonstrated by grants, fellowships, patents, as well as first-authored publications at leading workshops or conferences such as NeurIPS, ICLR, AAAI, RecSys, KDD, IJCAI, CVPR, ECCV, ACL, NAACL, EACL, ICASSP, or similar

Compensation: $88.46/hour to $257,000/year + bonus + equity + benefits