AI Evaluation & Benchmarking Lead — Data Scientist

Pwc; Leeds; Hybrid; GBP 90,000 - 130,000; Full time

Benefits offered by this job

empowered flexibility

private medical cover

volunteering days

Job summary

PwC Leeds is seeking a Data Scientist - AI Evaluation & Benchmarking Manager to shape model benchmarking and experimentation capabilities. You will drive evidence-based decisions for client engagements and own scalable AI research infrastructure in a fast-moving, collaborative environment.

You will lead end-to-end benchmarking, design workflows, and produce client-ready insights while mentoring others and advancing technical strategy across PwC projects.

Qualifications

• Hands-on data science or LLM experimentation experience.
• Proficient in Python and scalable code practices.
• Experience deploying ML workloads to cloud platforms and containerisation.
• Knowledge of statistics and experimental design.
• Ability to lead and influence technical direction.
• Autonomous, able to manage fast-moving workstreams.

Responsibilities

• Build and evolve AI benchmarking and experimentation platforms.
• Influence AI model selection and technical strategy across PwC projects.
• Design and run end-to-end benchmarking workflows.
• Build scalable evaluation frameworks and pipelines.
• Develop, maintain and improve experimentation infrastructure.
• Provide client-ready insights and support product demos.

Skills

Python

Asynchronous programming

Multithreading

Data science concepts

Experiment design

Leadership

Communication

Tools

Docker

Podman

Azure

AWS

GCP

CI/CD

Job description

PwC Leeds is seeking a Data Scientist - AI Evaluation & Benchmarking Manager to shape model benchmarking and experimentation capabilities. You will drive evidence-based decisions for client engagements and own scalable AI research infrastructure in a fast-moving, collaborative environment.

You will lead end-to-end benchmarking, design workflows, and produce client-ready insights while mentoring others and advancing technical strategy across PwC projects.