AI Orchestration Engineer

Design the nervous system of our AI products - multi-agent pipelines, tool-calling, RAG and eval loops that run in production. WHAT YOULL DO Architect multi-agent and LLM orchestration pipelines: routing, tool-calling, retrieval, guardrails and fallbacks. Build RAG systems over enterprise data with grounding, citations and auditability. Design evaluation harnesses and monitoring so model quality is measured continuously, not assumed. Optimise latency and cost across model providers; own pre-computation and caching strategies. Work directly with clients domain experts to turn tribal knowledge into agentic workflows. WHAT WE LOOK FOR 3+ years backend engineering (Python preferred); production experience with LLM APIs or frameworks. Hands-on with vector stores, embeddings and prompt/context engineering. You think in systems: retries, idempotency, observability come naturally. .