Staff AI Engineer — Self-Improving Engine Systems

Hone; Witney; On-site; GBP 113,000 - 174,000; Full time

Benefits offered by this job

Health and dental insurance

Maternity and parental support

Relocation support

Job summary

Hone is building AI engines that own outcomes, not tasks, with agents that learn and improve over time. This role focuses on offline evals, simulations, memory, and self-improvement loops that translate production feedback into updated skills, memory, code and policies.

You will work on memory systems, grading trajectories, and long-horizon evaluations that shape practical agent behavior in real companies.

Qualifications

• 2+ years of engineering experience or equivalent depth from research/open source.
• Experience at a fast-growing startup.
• Shipped applied ML systems from scratch as an engineer.
• Ran evals and experiments rigorously against production systems.
• Strong fundamentals: ML evaluation, experimental design, statistics, LLM agents.
• You use coding agents daily and still understand every line you ship.
• Built and tinkered for fun because you care how the tech works.

Responsibilities

• Simulations built from company context with simulated users and connectors to evaluate engines on tasks that take weeks or months.
• Turning production traces into simulations that we can rewind, fork with one change, and run forward, keeping the simulated world consistent.
• Memory systems that store, retrieve, and compact what engines learn.
• Self-improvement loops that turn traces into verified changes to skills, memory, code and policies.
• Grading and credit assignment over month-long trajectories.

Skills

Engineering experience

Startup experience

Applied ML systems

Experimentation & evals

ML evaluation

Experimental design

Statistics

LLM agents

Coding agents

Curiosity / tinkering

Lab AI experience

Research publications

Job description

Hone is building AI engines that own outcomes, not tasks, with agents that learn and improve over time. This role focuses on offline evals, simulations, memory, and self-improvement loops that translate production feedback into updated skills, memory, code and policies.

You will work on memory systems, grading trajectories, and long-horizon evaluations that shape practical agent behavior in real companies.