PhD/Postdoc physicists to design and solve challenging Physics problems that evaluate the limitations of large language models — from undergraduate to advanced PhD-level topics. At least 4 hours/day, up to 40 hours/week, with 4-hour PST overlap.
Physics Specialist
Remote (Global)
June 14, 2026
Job description
About the Role
Turing is seeking a Physics Specialist (Scientific Reasoning & Discovery Engineer) to help evaluate and improve the scientific reasoning capabilities of advanced AI systems.
In this role, you will create physics-focused reasoning datasets that challenge Large Language Models (LLMs) to analyze experimental and simulated observations, identify patterns, infer governing laws, estimate parameters, validate hypotheses, and predict outcomes.
The position combines physics expertise, mathematical modeling, quantitative analysis, and scientific reasoning to help build more capable AI systems for scientific discovery and problem solving.
Key Responsibilities
Design Scientific Reasoning Tasks
- Create reasoning scenarios using experimental datasets, observational data, and simulated physical systems
- Develop tasks requiring data analysis, pattern recognition, parameter estimation, predictive reasoning, and scientific inference
Build Scientific Evaluation Problems
- Create evaluation challenges focused on law discovery, model selection, hypothesis validation, consistency checking, and scientific reasoning
- Design tasks that require multi-step analytical thinking and quantitative problem solving
Develop Ground Truth Solutions
- Create deterministic answers and reference solutions
- Write detailed scientific explanations
- Build comprehensive scoring rubrics for model evaluation
Support AI Research
- Collaborate with reviewers and AI engineers
- Ensure datasets are scientifically accurate, logically consistent, reproducible, and clearly documented
Required Qualifications
- 3+ years of experience in Physics or a related scientific field
- Strong understanding of scientific reasoning, quantitative analysis, experimental methodology, and data interpretation
- Expertise in mathematical modeling, parameter estimation, dimensional analysis, and simulation-based problem solving
- Strong scientific writing and documentation skills
- Exceptional attention to detail and logical reasoning
Work Expectations
- Minimum overlap of 4 hours with PST time zone
- Fully remote work environment
- Contractor engagement
- Duration of approximately 8 weeks
Benefits
- Fully remote work
- Opportunity to work on cutting-edge AI projects
- Collaboration with leading LLM companies
- Potential contract extension based on performance and project needs
About Turing
Turing is a leading AI research accelerator that supports frontier AI laboratories and global enterprises through advanced training data, research expertise, and AI system development.
Apply now
You will be redirected to the company's website to complete your application.
Please submit your CV in English and indicate your level of English proficiency. Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems.
Please submit your CV in English and indicate your level of English proficiency. Mindrift connects specialists with project-based AI opportunities for leading tech companies, focused on testing, evaluating, and improving AI systems.
Design graduate-level computational problems using astropy and related tools for cosmological calculations, angular power spectra, and galaxy survey analysis — calibrating tasks against frontier AI models to build advanced AI reasoning benchmarks.