AI Agent Developer

oOlie AI; Greater London; On-site; GBP 70,000 - 110,000; Full time

Job summary

oOlie AI is seeking an experienced engineer to design agent systems capable of reasoning over structured and unstructured business context. You will build workflows that orchestrate tool use, retrieval, memory, and handoffs, and you will implement evaluation harnesses to ensure reliability, safety, and business correctness.

You will collaborate with product engineers to embed agents into production interfaces, focusing on latency, cost efficiency, and robust failure recovery under real user

Qualifications

• Experience building LLM or agent products beyond prototypes.
• Strong TS or Python skills.
• Practical understanding of retrieval, tool calling, and prompt/system design.
• Ability to debug complex behavior across distributed systems and model layers.

Responsibilities

• Design agent workflows spanning tool use, retrieval, memory, and handoffs.
• Build evaluation harnesses for reliability, safety, and business correctness.
• Collaborate with product engineers to embed agents into production interfaces.
• Improve latency, cost efficiency, and failure recovery under real user traffic.

Skills

TypeScript

Python

Retrieval

Tool calling

Prompt design

System design

Distributed systems debugging

Education

BSc in Computer Science

BSc in Software Engineering

BSc in AI

Tools

Production interfaces

Job description

Role Overview

You'll design agent systems that can reason over structured and unstructured business context, call tools safely, recover from failure, and generate outputs that are trustworthy in enterprise settings.

This is not prompt-only work. You'll be working across orchestration, evaluation, retrieval, tool contracts, state, and product behavior to create agents that are useful in production, not just impressive in demos.

What You'll Own

• Design agent workflows spanning tool use, retrieval, memory, and handoffs.
• Build evaluation harnesses for reliability, safety, and business correctness.
• Work with product engineers to embed agents into production interfaces.
• Improve latency, cost efficiency, and failure recovery under real user traffic.

What We're Looking For

• Experience building LLM or agent products beyond prototypes.
• Strong TypeScript or Python skills.
• Practical understanding of retrieval, tool calling, and prompt/system design.
• Ability to debug complex behavior across distributed systems and model layers.
• BSc in Computer Science, Software Engineering, AI, or equivalent applied experience.