Lead Applied AI Scientist: Build & Deploy Scalable ML

JPMorgan Chase & Co.; Greater London; On-site; GBP 90,000 - 130,000; Full time

Job summary

J.P. Morgan is seeking an Applied AI Scientist to join the AI Technologies team.

The role focuses on turning experimental ML and generative AI into scalable, production-ready solutions built with cross-functional collaboration across product, data engineering, software engineering, and other ML practitioners. The candidate will define problems, design experiments, and own Python development from prototypes to tested production code, balancing model quality, latency, and cost while mentoring

Qualifications

• Minimum 5 years of applied ML experience.
• Strong ML/deep learning knowledge and evaluation.
• Proficient Python and maintainable code.
• Hands-on experience with LLM-based systems.
• Ability to translate unclear requirements into testable problems.
• STEM degree or equivalent.

Responsibilities

• Define problems, success measures, and technical approaches with product and engineering teams.
• Design experiments, establish baselines, and benchmark models and agent systems.
• Own Python development from experiments to production code.
• Explain trade-offs between model quality, reliability, latency, and cost.
• Lead technical design reviews and mentor others.
• Work with teams to meet data, governance, and security requirements.

Skills

Applied ML

Python

LLM systems

Experiment design

Independent work

Communication

Education

STEM degree

Tools

PyTorch

TensorFlow

AWS SageMaker

Java/C/C++

Job description

J.P. Morgan is seeking an Applied AI Scientist to join the AI Technologies team.

The role focuses on turning experimental ML and generative AI into scalable, production-ready solutions built with cross-functional collaboration across product, data engineering, software engineering, and other ML practitioners. The candidate will define problems, design experiments, and own Python development from prototypes to tested production code, balancing model quality, latency, and cost while mentoring