Senior AI/ML Engineer — Autonomous Coding Agent

JetBrains; Greater London; Remote; GBP 80.000 - 120.000; Vollzeit

Benefits dieser Stelle

Flexible work location

Remote work abroad up to 30 days ayear

Medical insurance allowance

Learning and development opportunities

Relocation support

Language classes

Lunch allowance

Mental health support

On-site gym stipend

Internal events

Extra time off

Zusammenfassung

JetBrains is hiring to advance Junie, our AI coding agent available across IDEs and as a standalone CLI. You will own end-to-end features from research to production within three focus areas: CLI independence, model ensembling optimization, and high-autonomy subsystems.

You will collaborate with talented engineers to ship robust solutions, experiment with costs, and continually improve the tool's quality and usefulness for developers.

Qualifikationen

• Proven experience shipping ML or full-stack software products.
• Hands-on with LLMs, prompt engineering, or agentic workflows.
• Ability to write production-ready code across the full lifecycle.
• Experience building or fine-tuning ML models.

Aufgaben

• Develop and ship new features for Junie across CLI and model integrations.
• Experiment with model setups, ensembling, and cost efficiency.
• Build and test autonomous agent behaviors for long-term project maintenance.
• Own the full development cycle from initial research to production-ready code.
• Use Junie daily to improve quality and user experience.

Kenntnisse

ML/AI systems

LLMs

Prompt engineering

Agentic workflows

Production code

Full-stack software

Ausbildung

Bachelor's degree in Computer Science or related field

Tools

Python

PyTorch

CLI tools

Jobbeschreibung

JetBrains is hiring to advance Junie, our AI coding agent available across IDEs and as a standalone CLI. You will own end-to-end features from research to production within three focus areas: CLI independence, model ensembling optimization, and high-autonomy subsystems.

You will collaborate with talented engineers to ship robust solutions, experiment with costs, and continually improve the tool's quality and usefulness for developers.