Production LLM Agent Engineer — Reliability & Adoption

Obsidian; Greater London; On-site; GBP 89,000 - 133,000; Full time

Job summary

Obsidian is seeking engineers to build and operate production-grade LLM agents with real visibility into how they are used inside a company. You will work on internal monoagents wired into company data, reusable skills and playbooks, and the surfaces agents call.

Expect to discuss agent reliability, evaluation, and adoption with the team and share war stories from real systems. The process begins with a short conversational AI interview.

Qualifications

• Shipped an agent that real users depended on and been on the hook when it broke.
• Figured out how to tell whether an agent got better or worse after a change.
• Run agents that outlive a single request: scheduled jobs, long-running work, cloud sandboxes.

Responsibilities

• Build and operate production-grade LLM agents in a corporate setting.
• Evaluate agent reliability, performance, and adoption across teams.
• Analyze costs, usage, and visibility of agent activity within the company.

Skills

Agent reliability

Production deployment

Shipping agents

Evaluation methods

Long-running jobs

Job description

Obsidian is seeking engineers to build and operate production-grade LLM agents with real visibility into how they are used inside a company. You will work on internal monoagents wired into company data, reusable skills and playbooks, and the surfaces agents call.

Expect to discuss agent reliability, evaluation, and adoption with the team and share war stories from real systems. The process begins with a short conversational AI interview.