Founding AI Engineer (+ Equity) at OpenHuman

Job Title

Founding AI Engineer

Salary

Not Disclosed + Equity

Company Description

OpenHuman is an early-stage London startup building a category-defining decision platform that helps companies preview the future. By creating auditable AI systems for reasoning about uncertainty, OpenHuman enables teams to test pricing, product, and growth decisions in a safe environment before they are deployed in the real world.

Job Description

OpenHuman is building a decision platform that enables companies to preview the impact of high-stakes product and pricing changes before they go live. As a Founding AI Engineer, you will design the intelligence layer, building auditable LLM and agentic workflows that reason over messy data to provide testable, evidence-based forecasts.

Location

London, UK

Why this role is remarkable

• Shape a category-defining AI product from the ground up, moving beyond chatbots to build a sophisticated decision intelligence platform.
• Solve deeply complex technical challenges involving reasoning under uncertainty, counterfactual analysis, and creating auditable AI systems that separate evidence from speculation.
• Join an early-stage team in London with direct founder access and high ownership over technical architecture and product direction.

What You Will Do

• Design and build production-grade LLM and agentic workflows that transform customer, market, and operational data into reliable reasoning contexts.
• Develop rigorous evaluation and quality-control systems for nondeterministic AI outputs, including tracing and replay frameworks for model comparison.
• Collaborate with founders to define how the platform measures prediction accuracy and trustworthiness against real-world business outcomes.

The ideal candidate

• Brings 5+ years of software engineering experience with deep expertise in Python and recent hands-on work building production-level LLM or RAG systems.
• Possesses an evaluation-driven mindset, focusing on model failure modes, hallucination management, and statistical calibration of AI outputs.
• Demonstrates high agency in ambiguous startup environments, with the ability to turn customer problems into repeatable, production-ready AI workflows.

#J-18808-Ljbffr…