Computational Scientist – AI & Materials

Quantum Formatics; Cambridge; On-site; GBP 89,000 - 134,000; Full time

Benefits offered by this job

Hybrid work environment at The Engine

Competitive salary based on experience

Health, dental, and vision insurance

401(k)

Unlimited PTO

Relocation reimbursement

Job summary

Quantum Formatics is seeking a Computational Scientist to advance an AI-accelerated materials discovery pipeline. You will develop graph neural networks for materials property prediction and integrate ML interatomic potentials while benchmarking against experimental data.

You will collaborate with the Lead Scientist and experimental teams, mentoring future members as the team grows, and help shape the computational platform direction.

Qualifications

• Ph.D. with a strong focus on atomistic modeling or AI-accelerated materials discovery.
• At least 3 years of research experience in AI for materials discovery or related field.
• Experience with graph neural networks for materials property prediction.
• Familiarity with DFT and MD simulations; proficiency in Python and PyTorch.
• Experience training ML models for scientific applications and analyzing large datasets.
• Ability to work in HPC environments and manage computational workflows at scale.

Responsibilities

• Develop and train graph neural network models for materials property prediction.
• Evaluate and integrate machine learning interatomic potentials; benchmark against experimental data.
• Collaborate with the Lead Scientist and experimental teams to close the loop prediction–synthesis–characterization–model improvement.

Skills

Graph neural networks

Python

PyTorch

Density functional theory

Molecular dynamics

HPC workflows

Data analysis

Collaboration skills

Education

Ph.D. in Physics, Chemistry, Materials Science, Computer Science, or closely related field

Tools

Fortran

C

Job description

About the Role

We are seeking a Computational Scientist to join our R&D team and contribute directly to the continued development of our AI-accelerated superconductor discovery algorithm. You will implement methods for direct structure property predction.

Your day-to-day work will span developing and training graph neural network models for materials property prediction, evaluating and integrating state-of-the-art machine learning interatomic potentials, and developing benchmarks to validate model performance against experimental data. You will collaborate closely with our Lead Scientist, as well as with experimental collaborators who synthesize and characterize the materials your models identify. That loop, from prediction to synthesis to characterization and back into model improvement, is the core of what we do.

This is a foundational hire. As the team and the discovery pipeline grow, this role has a clear path toward technical leadership, with the opportunity to shape the direction of our computational platform and mentor and lead future team members.

About you (strongly preferred)

• Ph.D. in Physics, Chemistry, Materials Science, Computer Science, or a closely related field, with a strong focus on atomistic modeling, AI-accelerated materials discovery, or condensed matter physics or relevant research experience.
• At least 3 years of research experience in AI for materials discovery, computational materials science, or a closely related domain.
• Experience with graph neural networks for materials property prediction.
• Familiarity with density functional theory and molecular dynamics simulations.
• Experience training and evaluating machine learning models for scientific applications, with proficiency in Python and PyTorch.
• Familiarity with high-performance computing environments and managing computational workflows at scale.
• Ability to analyze and interpret large, complex datasets generated from simulations and experiments
• Excellent collaboration and communication skills, with the ability to articulate complex technical ideas clearly, and work effectively in a small, interdisciplinary team.

Pluses

• Familiarity with equivariant architectures (e.g., e3NN).
• Familiarity with Quantum Espresso or similar DFT codes for electron-phonon calculations and phonon property prediction.
• Experience with machine learning interatomic potentials (MLIPs)
• Experience with generative models for crystal structure generation (flow matching, diffusion models, or related architectures).
• Understanding of BCS theory, Eliashberg theory, and the physics of electron-phonon superconductors.
• Experience in Fortran and/or C .

What We Offer

• Real agency and real impact. At our size, every team member shapes the trajectory of the company. You will work directly with the CEO and CSO on decisions that define our discovery roadmap. The materials you help identify will move from your simulations to synthesis to commercial products that power fusion reactors, MRI systems, and energy infrastructure.
• A genuinely collaborative, mission-driven team. We are scientists who care about impact, not publicity. We work in an environment where ideas are heard, collaboration is the default, and everyone has a stake in the outcome.
• A flexible, hybrid work environment based at The Engine, MIT’s tough tech incubator, located in the heart of Kendall Square. Easily accessible by public transit and surrounded by one of the most vibrant innovation communities in the world.
• Competitive salary based on experience level and cost-of-living in Cambridge, MA, plus meaningful equity ownership.
• Employee sponsored benefits including health, dental, and vision insurance, plus a 401(k).
• Unlimited paid time off (PTO): minimum three weeks, federal holidays, and a company-wide closure between Christmas and New Years to support work-life balance.
• Relocation reimbursement for candidates moving to the Boston/Cambridge area.