Remote Pod Lead - Physics

Turing, one of the world’s fastest-growing AI companies, is dedicated to accelerating the advancement and deployment of powerful AI systems. The company partners with leading AI labs to push the boundaries of model capabilities in areas such as reasoning, coding, multimodality, multilinguality, and STEM knowledge. Leveraging these advancements, Turing builds real-world AI solutions that address critical business priorities.

We are currently seeking an experienced Physics expert to lead a team of trainers focused on developing realistic, terminal-based scientific tasks for Terminal Bench Science. In this role, you will be responsible for ensuring the physical accuracy, reproducibility, and reliability of every task produced by your team. You will provide coaching and mentorship to maintain a high standard of quality and challenge in the tasks delivered.

Key Responsibilities:
- Conduct thorough reviews of trainer-created tasks, including task statements, datasets, simulation inputs, environments, reference solutions, and automated tests.
- Ensure tasks authentically represent physics workflows and require genuine multi-step reasoning.
- Verify that tools, libraries, and dependencies are properly pinned to guarantee reproducibility of results.
- Validate dimensional consistency, unit systems, conservation laws, symmetries, integrator energy drift, convergence, and statistical treatment of uncertainties.
- Confirm that tolerances reflect legitimate stochastic and numerical variations without allowing physically incorrect outcomes.
- Oversee grader accuracy by ensuring they verify observables, fitted parameters, uncertainties, or physical invariants rather than superficial formatting.
- Prevent tasks from being gamed through hard-coded outputs or bypassed computations.
- Provide clear, actionable written feedback and manage task revisions through to completion.
- Onboard, mentor, and support trainers to foster continuous improvement.
- Allocate work effectively, monitor quality and throughput, and escalate any blockers promptly.
- Maintain comprehensive pod documentation, including guidelines, recurring issues, and examples of exemplary tasks.

Required Qualifications:
- Ph.D., postdoctoral experience, or equivalent advanced experience in physics.
- Strong programming skills in Python, C/C++, Julia, or Fortran, with proficiency in Linux environments.
- Hands-on experience in computational physics, simulations, or physics data analysis.
- Proven track record of reviewing code, analyses, or research outputs critically.
- Experience mentoring or leading small technical or research teams.
- Excellent written communication skills, capable of delivering precise and constructive feedback.

Preferred Qualifications and Benefits:
- Familiarity with scientific computing tools such as QuTiP, Qiskit, Kwant, ITensor, ROOT, Geant4, Astropy, Dedalus, or MEEP.
- Experience with Docker, Git, continuous integration systems, and automated testing frameworks.
- Background in designing benchmarks or simulation-based evaluations.
- Bonus points for experience evaluating AI coding or terminal agents, working in large collaborations with formal internal review processes, or prior involvement in AI training or evaluation programs.

Offer Details:
This is a contractor assignment requiring a commitment of at least 4 hours per day and a minimum of 40 hours per week, with a mandatory 4-hour overlap with Pacific Standard Time (PST). The contract duration is 4 weeks, with an expected start date as early as next week. Please note that this role does not include medical benefits or paid leave.

This opportunity is ideal for a physics professional passionate about applying their expertise to cutting-edge AI training and evaluation, leading a dedicated team to deliver scientifically rigorous and challenging tasks.