AI Researcher: Reasoning & Agency for Scientific Simulation

Mirror Physics • New York City • Full-time

$120,000

per year

Machine Learning Physics Reinforcement Learning AI Researcher Reasoning Scientific Simulation Quantum Chemistry

Job Description

The Company

Mirror Physics is a New York-based AI company working on a new frontier in scientific simulation. We design intelligent systems that understand physics from first principles, providing critical acceleration for advanced technological R&D. Today, we’re building the world’s most capable AI platform for predicting experimental outcomes in chemistry and materials science, tightly coupled with reality via high-throughput experimental verification, to accelerate discovery in biotech, energy, manufacturing, and other domains. Backed by leading investors and scientific experts, we are expanding our research team at a pivotal moment in the field.

The Opportunity

World-class physics models are only as powerful as the workflows that steer them. As the lead on Mirror’s reasoning team, your role is to design and engineer intelligent systems that understand how to run scientific simulation frameworks such as quantum chemistry and molecular dynamics, interpret the results, and adapt future actions. Your work will unlock autonomous R&D loops that plan experiments, allocate compute, and derive insights in real time, enabling state-of-the-art simulation methods to be widely and efficiently applied to industrial-scale problems.

Key Responsibilities

Architect LLM-based intelligent agents that plan, schedule, and monitor multistep simulation campaigns (DFT, ab-initio MD, reactive force-field, continuum).
Design reinforcement-learning or curriculum-learning loops that teach agents to balance exploration, accuracy, and cost across heterogeneous compute resources.
Integrate domain-specific knowledge graphs, symbolic reasoning engines, and uncertainty estimators.
Develop analytic tooling that explains agent decisions, determines root causes for simulation failures, and quantifies downstream business impact.
Build robust orchestration pipelines (Ray/Kubernetes/SLURM) for on-prem HPC and cloud GPU clusters; implement fault tolerance and provenance tracking.
Collaborate with foundation-model and multimodal teams to couple high-level reasoning with foundational atomistic predictions.
Engage with the AI-for-science community through publications and contributions at NeurIPS, ICML, ICLR, or other domain venues.

Who you are

Ph.D. or M.S./B.S. with equivalent research record in Computer Science, Applied Math, or related field with emphasis on reasoning and LLMs.
3+ years research experience building decision-making or agentic frameworks at scale.
Fluency in Python plus modern ML stacks (PyTorch/JAX) and familiarity with distributed training tooling (CUDA, NCCL, Slurm/K8s/Ray).
Publications in top-tier ML or domain conferences/journals.
Strong publication or open-source track record in ML for physical sciences.
Excellent collaboration, communication, and team-working skills.
Deep commitment and passion for advancing science.

Preferred Extras

Familiarity with quantum chemistry, atomistic simulation, or chemistry/materials science
Familiarity with active learning, retrieval-augmented generation, or agentic workflows for scientific automation.
Prior experience aligning language models with scientific knowledge bases or ontologies.

What We Offer