Model Behavior Architect (AI)

TL;DR

Model Behavior Architect (AI): Defining and measuring how LLMs use tools, invoke functions, and orchestrate agentic workflows with an accent on model evaluation and eval pipelines. Focus on designing synthetic tool environments, improving schema adherence, and solving complex multi-step tool orchestration challenges.

Location: Hybrid in London or Paris

Company

A pioneering AI company democratizing high-performance, open-source models and solutions for enterprise and personal use.

What you will do

• Identify and improve function calling and tool use behavior through direct model interaction.
• Design and implement evaluations, data guidelines, and synthetic tool environments and APIs.
• Resolve edge case behaviors, including malformed arguments, hallucinated functions, and incorrect tool selection.
• Develop robust evaluation pipelines for the function-calling capabilities of model candidates.
• Collaborate closely with AI Scientists to define quality standards for tool orchestration and reasoning.

Requirements

• Deep understanding of API design, structured outputs, and schema specifications (e.g., JSON Schema).
• Expertise in engineering, code behavior, or LLM agents, including reasoning and planning.
• Prior experience in training and optimizing model behavior.
• Proven track record of building robust evaluations and delivering innovative solutions.
• Ability to thrive in dynamic and technically complex environments.

Culture & Benefits

• Competitive salary and equity (stock options).
• Comprehensive health insurance.
• Transportation and sport allowances.
• Meal vouchers.
• Generous parental leave policy.
• Visa sponsorship.