Founding AI Engineer - LLM Pipelines & Systems

HartleyCo; England; On-site; GBP 120,000 - 160,000; Full time

Job summary

HartleyCo in the United Kingdom is seeking a founding engineer to build the agent systems that bridge specs to shipped AI features. You will design, build and operate the pipeline from spec to shipped build, working directly with the founder and founding engineering team to bring product ideas to life.

You will create evaluation tooling, routing layers, and context/inference cost optimizations, with broad scope to influence how AI is used across the product in an in-person, high-ownership

Qualifications

• Experience with production-grade LLM systems and multi-step pipelines.
• Experience designing and running evals with test sets and benchmarks.
• Strong systems engineering background; TypeScript preferred.
• Proven ability to optimize inference cost while maintaining quality and latency.
• Willingness for in-person, high-ownership startup environment.

Responsibilities

• Design, build and operate the pipeline from spec to shipped build: planning, generation, review and release.
• Build evaluation and regression tooling to catch quality issues before they ship.
• Build the routing layer that selects the right model for each step on quality, latency and cost.
• Own retrieval and context systems, with a path into fine-tuning models on the data the platform generates.
• Shape how AI is used across the whole product, not bolted onto one feature.

Skills

LLM pipelines

System design

TypeScript

Eval infrastructure

Cost optimization

Job description

HartleyCo in the United Kingdom is seeking a founding engineer to build the agent systems that bridge specs to shipped AI features. You will design, build and operate the pipeline from spec to shipped build, working directly with the founder and founding engineering team to bring product ideas to life.

You will create evaluation tooling, routing layers, and context/inference cost optimizations, with broad scope to influence how AI is used across the product in an in-person, high-ownership