Check location eligibility. Remote describes how the work is done; it does not necessarily mean applications are accepted worldwide. Check the location and residency requirements in this listing before applying.
About the role
Job Description: Computational Physical Scientist – AI Training & Model Evaluation
Role Type: Contractor
Location: Global (Fully Remote)
Time Commitment: 15+ hours per week
Schedule: Flexible
About the Role
We are seeking experienced Computational Physical Scientists with prior professional AI-training or model-evaluation experience to contribute to a project focused on improving advanced AI systems.
You will apply your scientific expertise and programming skills to develop challenging technical tasks, evaluate AI-generated code and analyses, identify errors, and establish objective standards for accurate solutions. This opportunity is intended for professionals with substantive AI-training experience involving physical-science content.
Key Responsibilities
Create realistic computational physical-science tasks that challenge advanced AI models.
Develop clear task instructions, reproducible environments, reference solutions, and evaluation criteria.
Evaluate AI-generated calculations, code, simulations, models, and technical deliverables.
Identify invalid assumptions, numerical errors, physically impossible results, incorrect methodologies, and reproducibility failures.
Assess the scientific accuracy, reliability, and completeness of model-generated outputs.
Establish objective standards for evaluating computational and scientific solutions.
Required Qualifications
Prior paid experience training, evaluating, benchmarking, or improving AI systems on scientific or technical tasks.
Strong practical proficiency in Python.
Deep expertise in at least one relevant physical-science specialization.
Hands-on experience using computational methods for modeling, simulation, numerical analysis, data analysis, or reproducible research.
Relevant Specializations
Physics
Astronomy and Cosmology
Materials Science
Chemistry
Computational Physics
Computational Chemistry
Scientific Simulation
Public Proof of Technical Expertise
Strong applicants should be able to demonstrate their expertise through evidence such as:
Peer-reviewed publications or preprints involving computational methods.
Public simulation or scientific analysis code.
Google Scholar, ORCID, arXiv, or equivalent research profiles.
Open-source scientific software contributions.
Reproducible notebooks, models, datasets, or simulations.
Public benchmarks, experiments, or technical projects.
Preferred Qualifications
Experience creating AI-training tasks, reference answers, evaluation rubrics, graders, or benchmark problems.
Familiarity with NumPy, SciPy, SymPy, Astropy, PyTorch, JAX, or comparable scientific tools.
Experience validating simulations or models against analytical results, observations, or experimental data.
Availability to begin immediately.
Application Process
Application Submission: Submit your application along with details of your AI-training experience, scientific specialization, and public technical evidence.
AI Interview: Complete an AI interview that includes technical and integrity checks.
Hiring Manager Review: The team evaluates your profile and interview performance.
Onboarding: Selected candidates complete project onboarding and begin their first assigned task.
Compensation
Compensation is output-based, with payment for tasks that meet project specifications and quality requirements. Task availability and completion time may vary.
Originally posted on Himalayas
