Postdoctoral Research Associate - Isayev Lab
The Isayev Lab at Carnegie Mellon University invites applications for a postdoctoral researcher to lead projects at the interface of computational chemistry, machine learning, reaction mechanism elucidation, and automated molecular discovery. The position is ideal for a candidate who wants to turn deep mechanistic understanding into predictive models and closed-loop discovery workflows.
Our lab develops and applies machine learning methods for computational chemistry, materials science, and molecular discovery, including transferable neural network potentials, generative molecular design, and experiment-automation workflows. The postdoc will work in a collaborative CMU environment spanning computational chemistry, AI, automated experimentation, polymer chemistry, and catalysis.
Research directions may include:
- Developing automated DFT / ML workflows for mechanistic studies of photoredox, organometallic, and radical catalytic reactions.
- Building predictive models that connect quantum-chemical descriptors, catalyst structure, substrate scope, selectivity, and reaction performance.
- Applying AIMNet2 and related ML/QM methods to accelerate conformer search, reaction-path exploration, catalyst screening, and high-throughput mechanistic modeling.
- Designing closed-loop computational-experimental campaigns for transition metal catalysis, polymer synthesis, and related catalytic transformations.
- Creating reusable, open, well-documented software workflows for reaction data generation, curation, featurization, and model deployment.
- Collaborating with experimental groups at CMU and external partners to convert mechanistic hypotheses into experimentally testable predictions.
Qualifications:
Desired background:
- Ph.D. in chemistry, chemical engineering, materials science, or a related field.
- Strong experience in computational reaction mechanisms, especially DFT studies of organic, organometallic, photoredox, radical, or homogeneous catalytic systems.
- Fluency with Python and modern scientific computing workflows; experience with Git, HPC clusters, SLURM, Gaussian, ORCA, Q-Chem, xTB, RDKit, ASE, or related tools is highly valued.
- Interest in machine learning, statistical modeling, active learning, descriptor development, or data-driven reaction prediction.
- Ability to work closely with experimental collaborators and communicate mechanistic insight clearly.
Find Your Best Opportunity
Tell them AcademicJobs.com sent you!
