Lead Pod for Mathematical Benchmarks and Task Review

Turing; California; Hybride; USD 150 000 - 200 000; Plein temps

Résumé du poste

Turing in the United States (California) seeks a lead for a pod of 5–10 trainers, crafting realistic terminal-based tasks for scientific benchmarks. You will review work end-to-end, verify correctness, and guide graders with rigorous error analysis.

You will mentor/trainers, manage throughput, and collaborate with stakeholders to improve benchmark quality and reproducibility across projects.

Qualifications

• PhD or equivalent in mathematics or related field.
• Strong programming and Linux skills.
• Experience mentoring or leading small technical teams.
• Ability to assess reproducibility and scientific validity of computations.

Responsabilités

• Lead a pod of trainers developing realistic, terminal-based tasks.
• Review trainer work for mathematical correctness and reproducibility.
• Mentor trainers; manage quality and throughput; escalate blockers.
• Design and review graders based on error bounds and proofs.
• Ensure tolerances and reproducibility; document processes.

Connaissances

Python

Julia

C/C++

MATLAB/Octave

Linux/Terminal

Numerical methods

Symbolic computation

Optimization

Computational statistics

Mathematical modeling

Formation

Ph.D. in mathematics / applied mathematics / statistics or related field

Outils

NumPy

SciPy

SymPy

mpmath

SageMath

PARI/GP

GAP

Macaulay2

Mathematica-like workflows

Docker

Git

CI / automated testing

PETSc

JuMP / Pyomo / CVXPY

Lean/mathlib or Coq

Isabelle

Description du poste

Turing in the United States (California) seeks a lead for a pod of 5–10 trainers, crafting realistic terminal-based tasks for scientific benchmarks. You will review work end-to-end, verify correctness, and guide graders with rigorous error analysis.

You will mentor/trainers, manage throughput, and collaborate with stakeholders to improve benchmark quality and reproducibility across projects.