Google : Mathematics and Finance Expert

About Turing:Based in San Francisco, California, Turing is the world’s leading research accelerator for frontier AI labs and a trusted partner for global enterprises deploying advanced AI systems. Turing supports customers in two ways: first, by accelerating frontier research with high-quality data, advanced training pipelines, plus top AI researchers who specialize in coding, reasoning, STEM, multilinguality, multimodality, and agents; and second, by applying that expertise to help enterprises transform AI from proof of concept into proprietary intelligence with systems that perform reliably, deliver measurable impact, and drive lasting results on the P&L. Role summaryTuring is seeking experts in mathematical finance to author and peer-review a curated set of challenging, well-specified quantitative problems. The goal is to produce rigorous, unambiguous items with numerical answers that create measurable headroom versus current frontier models—difficult for the right reasons (depth, novelty, and precision), not because of unclear wording or missing assumptions.

Key responsibilities

1. Author high-caliber mathematical finance questions • Write original, clearly specified problems (all assumptions and parameters defined; no ambiguity-by-design). • Target known model failure modes via multi-step derivations, parameter-sensitive reasoning, and less-common methods (e.g., recent or niche literature). • Ensure each problem has a checkable, numerically evaluable final answer (preferably with >2 significant digits to reduce guessing).
2. Include a “Help” component for every question • For each problem, provide a structured Help section that unlocks the solution path without revealing the final answer. • Help may include: required definitions, key lemmas, intermediate sub-questions, or a constrained hint ladder that makes the task solvable with targeted assistance.
3. Produce evaluation-ready solutions and participate in peer review • Deliver a full step-by-step solution that is logically complete, auditable, and reproducible. • Include sanity checks or limiting-case checks where relevant (e.g., parameter limits, dimensional analysis, monotonicity). • Peer review other experts’ questions for clarity, correctness, difficulty, and specification completeness.
4. Research-based question sets (recommended) • Select a recent or influential paper/method and develop a mini-set of 4–5 questions probing core ideas, assumptions, or derivations. • Design items that reward deep understanding (not memorization) and remain self-contained via the Help component.

Deliverables

• Problem statement with complete assumptions, definitions, and parameter values. • Numerical final answer (with reproducible computation; specify tolerance if needed). • Full solution (step-by-step derivation). • Help component (hint ladder / intermediate steps / key references). • Peer-review notes for assigned questions (accuracy, clarity, difficulty, and spec completeness).

Required expertise

• Graduate-level (or equivalent) mastery of mathematical finance. • Strong background in stochastic calculus (SDEs, martingales, Ito/Stratonovich, Girsanov/change of measure). • Derivative pricing across methods: PDE/BSDE approaches, Monte Carlo and variance reduction, calibration and implied volatility. • Experience with interest-rate and/or credit models (e.g., HJM/LMM/CIR/HW; reduced-form or structural credit). • Numerical methods literacy (finite differences, discretization error, stability, adjoints/Greeks, QMC where relevant). • Exceptional technical writing and notation hygiene; ability to make problems self-contained and evaluation-ready.

Preferred qualifications

• Familiarity with recent research directions (e.g., rough volatility, XVA, optimal execution/microstructure, advanced risk measures, robust finance). • Prior experience writing qualifying-exam-level or contest-quality problems with complete solutions and rubrics. • Comfort designing numerically stable answer keys (explicit tolerances, clear units/scales, and reproducibility). Quality bar • Difficulty must come from conceptual depth and rigor—not from missing data, trick phrasing, or hidden assumptions. • Problems must be self-contained: any specialized definition is included or introduced through the Help component. • Solutions must be independently checkable and numerically reproducible. AI-use policy (project constraint) Follow project-specific guidance on AI usage. When expert-only content is required, do not rely on external LLMs for content generation unless explicitly permitted by the project instructions. If limited AI assistance is allowed (e.g., for editing), use only the approved tool(s) and keep authorship and verification responsibilities with the expert.