Staff Applied AI Scientist

About the role

Culture Amp is seeking a Staff Applied AI Scientist to lead the observability and evaluation of its Coach AI system. You will build end‑to‑end feedback loops, design context‑engineering solutions, and create robust evaluation frameworks to ensure high‑performance production AI.

Key responsibilities

• Own the feedback loop: prompt engineering, large‑scale evaluation, and continuous improvement of LLM‑powered tools.
• Design and optimise context handling, retrieval, memory, and budget management for multi‑turn agentic flows.
• Develop and run evaluation pipelines using sampling, LLM‑as‑a‑judge, and de‑identified production traces.
• Guide model and provider selection balancing quality, latency, and cost.
• Create production guardrails for safety, PII handling, and content‑safety.

Required profile

• Proven experience building production agentic systems with context engineering, RAG, and memory management.
• Hands‑on expertise in LLM evaluation, observability tooling, and longitudinal performance measurement.
• Strong technical writing and ability to enable other teams through frameworks and documentation.

Required skills

• Langfuse or comparable LLM observability platform.
• LLM‑as‑a‑judge evaluation techniques.
• Agentic coding tools such as Claude Code, Cursor, or Codex.
• Prompt engineering, retrieval‑augmented generation (RAG), and context compression.

What we offer

• Equity through an Employee Share Option Program.
• Learning programs, coaching, and quarterly refresh days.
• Flexible hybrid work model (average 2 days onsite per week).
• MacBook and home‑office budget.
• Inclusive parental leave and wellbeing allowances.