AI Research Engineer — Production LLM Systems (London)

Edra; Greater London; On-site; GBP 134,000 - 194,000; Full time

Benefits offered by this job

Equity

Health insurance

Flexible PTO

Lunch in office

Workspace stipend

Job summary

Edra is building a learning system that turns customer data into actionable knowledge and instructions for AI agents. We seek AI Research Engineers who can scale LLM-based pipelines, develop production-grade orchestration, and push the boundaries of enterprise AI.

You will own components across context-learning, evaluation, and continuous learning loops. You’ll join a deeply technical team working in New York and London, contributing to knowledge benchmarks and the self-improving harness that

Qualifications

• 3+ years of professional experience in AI/ML roles.
• Experience building complex production LLM-based systems.
• Experience with context engineering, retrieval, or agent orchestration.

Responsibilities

• Build and extend our core context-learning library and reusable platform capabilities.
• Design and implement LLM-powered systems and agentic workflows from concept to production.
• Develop knowledge-management features: autonomous editing and maintenance of large knowledge bases.
• Create continuous learning loops using human feedback to identify knowledge gaps and improve prompts.
• Develop reliability and confidence systems: evaluation frameworks and decision logic for automation vs human input.
• Architect async, scalable systems that support complex AI orchestration.
• Contribute to the direction of our AI strategy and product capabilities.

Skills

Production LLM systems

Context engineering

Agent orchestration

Memory

Evaluation

Learning from feedback

Strong communicator

Experimental judgment

Open-ended problems

Job description

Edra is building a learning system that turns customer data into actionable knowledge and instructions for AI agents. We seek AI Research Engineers who can scale LLM-based pipelines, develop production-grade orchestration, and push the boundaries of enterprise AI.

You will own components across context-learning, evaluation, and continuous learning loops. You’ll join a deeply technical team working in New York and London, contributing to knowledge benchmarks and the self-improving harness that