Senior AI Agent Engineer — Production-Ready Systems (Remote)

Owen Thomas | B Corp™; Greater London; Hybrid; GBP 92,000 - 101,000; Full time

Benefits offered by this job

Remote work

Job summary

Agentic AI is seeking an AI Engineer focused on agentic systems to own the production agent stack, turning high-level requirements into reliable software. You’ll design agent loops, implement evals, manage context routing and optimize for quality, latency and cost while collaborating with founders and the engineering team.

You’ll contribute to building a startup-grade platform in the gaming space, with emphasis on robust code generation, tooling, and scalable model routing.

Qualifications

• Production experience building LLM/agentic systems.
• Multi-step pipelines, tool-using agents or autonomous systems.
• Agents that produce output that actually runs, not just text.
• Strong experience with evals, benchmarks and regression testing.
• Understanding trade-offs between model quality, latency and cost.
• Keeps pace with AI advances and experiments with new models.

Responsibilities

• Own the agent pipeline from specification to production.
• Build workflows across planning, code generation, review, testing and release.
• Build evals and regression tooling to measure output quality and catch failures.
• Develop model routing based on quality, latency and cost.
• Build retrieval and context systems.
• Improve or replace third-party agent infrastructure where needed.
• Experiment with new models, fine-tuning and post-training.

Skills

LLMs

Agentic systems

Evals

Langchain

Code generation

Model routing

TypeScript

Tools

Langchain

Langgraph

Retrieval Systems

Job description

Agentic AI is seeking an AI Engineer focused on agentic systems to own the production agent stack, turning high-level requirements into reliable software. You’ll design agent loops, implement evals, manage context routing and optimize for quality, latency and cost while collaborating with founders and the engineering team.

You’ll contribute to building a startup-grade platform in the gaming space, with emphasis on robust code generation, tooling, and scalable model routing.