About the Job
We’re big believers in the power of IRL, so for most roles we ask Campers to work from their local Culture Amp office an average of 2 days a week to unlock connection, pace and culture together.
Join us on our mission to make a better world of work.
Culture Amp is backed by leading venture capital funds and has offices in the US, UK, Germany and Australia. Culture Amp has been recognized as one of the world’s top private cloud companies by Forbes and most innovative companies by Fast Company.
Key Responsibilities
Design, implement and evaluate areas and features for agentic AI capabilities and systems using frameworks (e.g., LangGraph) for stateful, multi-turn conversations.
Advise, help build, and orchestrate data pipelines and integrations involving workplace data (surveys, goals, performance reviews, feedback, 1-1 notes) into coherent context view.
Build memory across sessions, and context graph, grounded in a custom ontology, alongside per-user relational and procedural memory and preferences that persist to the user profile.
Build and evaluate effective RAG, including pre-processing and hybrid search, and design areas of our graph-based agentic frameworks.
Own the end-to-end feedback loop: prompt engineering, evaluation at scale, and continuous improvement, including LLM-powered analysis tools that diagnose performance shifts and recommend prompt or system-level changes.
Translate customer requirements into technical solutions by working backwards from user needs to system design, and write the technical documentation that supports transfer to product teams.
Partner closely with product, design, and people science so features are fit for purpose and scale.
Contribute to evaluation frameworks and bias testing that meet enterprise requirements for transparency, fairness, and responsible AI.
Create and Monitor guardrails and safety in production.
Stay current with AI research, literature and provider offerings, and bring what's useful into production.
Required Skills & Abilities
Proven commercial experience taking ML/AI systems to production.
Hands-on experience with knowledge graphs and graph databases (e.g. Neo4j, Neptune).
Production LLM evaluation and observability: LLM-as-judge, eval datasets, human-in-the-loop labelling, scoring against thresholds, plus tracing and production monitoring (e.g. Langfuse, LangSmith).
Experience in designing and scaling production-grade multi-agentic systems (orchestration, prompt and context engineering, RAG, memory, agent architectures)
Full stack software development experience, ideally Python, React, PGVector
Strong software engineering fundamentals - you write clean, tested, maintainable code.
Proven ability to instruct and delegate complex workflows to autonomous agents, building the necessary feedback loops to ensure they act reliably on your behalf.
Hands-on experience designing and integrating data pipelines - you're comfortable moving data across systems, working within rigorous security standards, handling messy real-world inputs, and building robust orchestration
You are ambitious with what you build whilst maintaining a pragmatic, solution-focused mindset - you evaluate new technology through the lens of "does this scalably solve the problem?" not "is this technically interesting?"
Values alignment, the initiative to take risks, and the humility to learn and work out loud. You take pride in your work and its successful utilisation, stability, extendability and maintainability.
Experience at startups or founding teams - comfortable with ambiguity and fast iteration
Postgraduate degree (Masters or PhD) in Machine Learning, Computer Science, Applied Mathematics, or related quantitative field
Evidence of taking cutting-edge (relevant) techniques from paper to production
Experience with evaluation/benchmarking frameworks for AI systems
Experience implementing GraphRAG or Knowledge Graphs (e.g., Neo4j, Neptune)
Evidence of taking relevant, recent techniques from paper to production
Familiarity with people analytics, HR tech, or behavioural science applications
Open source contributions or public technical writing
