---
title: "Research Scientist, AI for Mathematics & Physical Sciences"
slug: "research-scientist-ai-for-mathematics-physical"
id: "1ced433d-4943-4102-bf8b-443e60e08e0b"
department: "AI Research"
department_code: "RES"
location: "Cambridge, UK"
workplace_type: "Hybrid"
employment_type: "Full-time"
compensation: "£130k – £185k (annual package)"
is_remote: false
is_featured: false
headcount: 1
canonical_url: "https://jobs.talent247.co/jobs/research-scientist-ai-for-mathematics-physical"
markdown_url: "https://jobs.talent247.co/jobs/research-scientist-ai-for-mathematics-physical.md"
apply_endpoint: "https://jobs.talent247.co/api/public/jobs/research-scientist-ai-for-mathematics-physical/apply"
schema_endpoint: "https://jobs.talent247.co/api/public/jobs/research-scientist-ai-for-mathematics-physical/form-schema"
webmcp_tool: "apply_for_role"
updated_at: "2026-11-15T17:32:05.601Z"
content_signal: "search=yes, ai-input=yes, ai-train=no"
---

# Research Scientist, AI for Mathematics & Physical Sciences

## Position Telemetry

- **Reference:** `RESEARCH-SCIENTIST`
- **Practice / Sector:** AI Research (`RES`)
- **Location:** Cambridge, UK
- **Workplace Type:** Hybrid
- **Employment Type:** Full-time
- **Compensation:** £130k – £185k (annual package)
- **Canonical HTML URL:** https://jobs.talent247.co/jobs/research-scientist-ai-for-mathematics-physical

## Role Dossier & Specification

THE CLIENT & MANDATE
Our client is a premier frontier AI research laboratory expanding their interdisciplinary scientific discovery hub in Cambridge, UK. Their AI for Science & Formal Reasoning division applies foundation models, geometric deep learning, and reinforcement learning to solve grand challenges in pure mathematics, quantum chemistry, materials discovery, and plasma physics.

They have retained Talent247 to appoint a Research Scientist in AI for Mathematics & Physical Sciences in Cambridge. Collaborating with Fields-caliber mathematicians, theoretical physicists, and ML systems engineers, the appointed scientist will develop neuro-symbolic theorem provers and physically constrained neural surrogates that discover verifiable scientific insights.

KEY RESPONSIBILITIES
• Formulate novel machine learning architectures and reinforcement learning search methods for formal mathematical reasoning and automated theorem proving in interactive proof assistants such as Lean 4.
• Develop equivariant geometric neural networks, neural operators, and score-based generative models for quantum many-body systems, crystalline materials, and partial differential equations (PDEs).
• Build closed-loop synthetic conjecture generation and formal verification curricula that overcome human-proof data scarcity through self-play and verifier feedback.
• Collaborate with academic and national-laboratory partners to validate model predictions against ab-initio DFT simulations, fusion plasma diagnostics, or formal proof libraries.
• Scale training and search inference experiments across multi-node GPU and custom accelerator clusters using JAX.
• Publish seminal results in leading machine learning and natural science journals while open-sourcing benchmarks that advance the scientific AI ecosystem.

TECHNICAL ENVIRONMENT & OFFERING
Stack: Python, JAX, PyTorch, Lean 4 / Isabelle, equivariant GNNs (e3nn), differentiable physics solvers, multi-node HPC/accelerator clusters. Workplace: Hybrid in Cambridge, UK. Compensation: £130,000 – £185,000 base salary plus equity/annual bonus, comprehensive benefits, and full UK Global Talent / Skilled Worker visa sponsorship.

## Core Requirements

ESSENTIAL QUALIFICATIONS
• PhD in Machine Learning, Mathematics, Theoretical Physics, Computational Chemistry, or Computer Science with a strong frontier research record.
• First-author publications at top ML conferences (NeurIPS, ICML, ICLR) or leading scientific journals (Nature, Science, PRL, JACS, Communications in Mathematical Physics).
• Deep proficiency in JAX or PyTorch, differentiable programming, and scaling neural architectures on distributed compute.
• Rigorous mathematical foundations in at least one target domain: formal logic/type theory, differential geometry, quantum mechanics, or numerical PDEs.
• Demonstrated ability to bridge abstract domain symmetries and constraints with clean, scalable deep learning implementations.

PREFERRED / BONUS EXPERIENCE
• Hands-on experience interfacing LLMs or tree-search RL agents with Lean 4 (Mathlib), Coq, or Isabelle proof kernels.
• Prior work on SE(3)/E(3)-equivariant neural networks, neural operators (FNO), flow matching on manifolds, or DFT/molecular dynamics surrogates.
• Olympiad mathematics/physics background or contributions to major open formalization or scientific ML repositories.

## Application Schema & Direct Apply (`apply_for_role`)

Candidates and authorized AI agents on behalf of a candidate can submit an application via the declarative WebMCP form tool `apply_for_role` on https://jobs.talent247.co/jobs/research-scientist-ai-for-mathematics-physical or via HTTP `POST https://jobs.talent247.co/api/public/jobs/research-scientist-ai-for-mathematics-physical/apply` (`application/json` or `multipart/form-data`). Non-browser HTTP agents must include header `X-Talent247-Intake-Token` obtained from `GET https://jobs.talent247.co/api/public/intake-token` (or automatically returned in `GET https://jobs.talent247.co/api/public/jobs/research-scientist-ai-for-mathematics-physical/form-schema`).

### Form Fields

- `full_name` (text, required): Full name
- `email` (email, required): Email
- `phone` (phone, optional): Phone
- `location` (text, optional): Location
- `linkedin_url` (url, optional): LinkedIn URL
- `cover_letter` (textarea, optional): Cover letter
- `consent` (checkbox, required): I consent to my data being processed for this application
- `cv_file` / `cv_base64` + `cv_filename`: Upload a CV in .pdf or .docx format (<= 10 MB) or provide an `https://` LinkedIn profile URL in `linkedin_url`.

### Alternative: Talent Pool Intake (`join_talent_pool`)

If this mandate is not an exact match, register once in the Talent247 Talent Pool at https://jobs.talent247.co/talent-pool (`https://jobs.talent247.co/talent-pool.md`) via WebMCP tool `join_talent_pool` or `POST https://jobs.talent247.co/api/public/talent-pool`.
