AI Research Scientist — Agentic AI for Materials Discovery at Meta

Redmond, Washington, United States -

Full Time

Start Date

Immediate

Expiry Date

11 Sep, 26

Salary

257000.0

Posted On

13 Jun, 26

Experience

2 year(s) or above

Remote Job

Yes

Telecommute

Yes

Sponsor Visa

Skills

LLM Orchestration, Multi-Agent Systems, Python, PyTorch, JAX, RAG, Computational Chemistry, Materials Informatics, DFT, Molecular Dynamics, HPC Infrastructure, Prompt Engineering

Industry

Software Development

Description

This role sits within Meta's Reality Labs Research, in the Material and Systems Innovation team, which develops advanced materials for two of Meta's most ambitious hardware frontiers: lightweight, all-day wearable AR glasses and next-generation sensing and actuating materials for robotics. The AI Specialist will design and build LLM-orchestrated multi-agent systems that autonomously drive materials discovery pipelines — from computational screening and simulation through synthesis and characterization — across both domains. By closing this loop with agentic AI, we aim to compress discovery timelines from years to weeks, directly accelerating Meta's ability to ship breakthrough AR/VR and robotics hardware. This position bridges two of Meta's highest-priority investment areas — frontier AI and the physical systems underpinning the metaverse and embodied intelligence — and will contribute both production systems and published research at top-tier venues. Responsibilities Design, implement, and optimize LLM-orchestrated multi-agent systems for autonomous materials discovery pipelines Build specialized AI sub-agents that operate within a closed-loop discovery framework Integrate agentic AI workflows with computational chemistry tools (DFT, MD, Monte Carlo) and HPC infrastructure Develop and fine-tune retrieval-augmented generation (RAG) systems over scientific literature corpora for real-time knowledge synthesis Collaborate with materials scientists, computational chemists, and ML researchers to translate domain workflows into autonomous agent architectures Evaluate and benchmark agent performance on materials discovery tasks — measuring accuracy, throughput, and synthetic viability of generated candidates Contribute to open-source tooling and publish research at top-tier venues (NeurIPS, ICML, ICLR, or domain journals) Minimum Qualifications Bachelor's degree in Computer Science, Computer Engineering, relevant technical field, or equivalent practical experience PhD in AI, Computer Science, Computational Chemistry, Materials Science, or related field 2+ years of experience with large language models, prompt engineering, or agentic AI frameworks (e.g., React, tool-use agents, multi-agent orchestration) Demonstrated programming skills in Python and experience with ML frameworks (PyTorch, JAX, or similar) Demonstrated experience in building end-to-end AI systems that integrate external tools and APIs Familiarity with at least one domain: computational chemistry, molecular simulation, or materials informatics Preferred Qualifications Experience building multi-agent or LLM-orchestrated systems for scientific applications Familiarity with atomistic simulation tools (VASP, Gaussian, LAMMPS, ASE) or cheminformatics libraries Publications at peer-reviewed ML or domain conferences Experience with retrieval-augmented generation, knowledge graphs, or scientific literature mining Understanding of crystal structure prediction, molecular dynamics, or quantum chemistry workflows Experience with HPC job orchestration $184,000/year to $257,000/year + bonus + equity + benefits

Responsibilities

Design and implement LLM-orchestrated multi-agent systems to automate materials discovery pipelines for AR glasses and robotics. Integrate AI workflows with computational chemistry tools and publish research at top-tier venues.