Agentic AI Engineer

Deliverance AI is building an Enterprise AI Operating System that provides governed, production-grade AI infrastructure to large regulated organisations in pharma, financial services, and telecoms. This role sits within the Forward Deployed Engineering team and focuses on designing and shipping production AI agent systems for enterprise clients — not research prototypes. Day-to-day work involves building multi-agent workflows, implementing RAG pipelines, configuring model routing, and ensuring every deployment meets strict governance requirements such as audit trails and PII redaction. The role offers meaningful influence over how autonomous AI operates inside large enterprises, with direct customer impact.

Responsibilities

• Design and deploy production AI agent systems for enterprise clients on the Deliverance AI platform
• Build multi-agent workflows using Swarm patterns, including orchestrators coordinating specialist agents
• Configure and extend agents from the platform library across multiple enterprise domains
• Implement RAG pipelines, model routing, and tool integrations for client-specific workflows
• Ensure all deployments meet Trust Architecture standards: audit trails, PII redaction, and human-in-the-loop gates
• Contribute novel agent patterns, prompt templates, and evaluation frameworks back to the platform

Requirements

• 4–8 years in software engineering, with 2+ years focused on AI/ML or LLM application development
• Strong Python skills and hands-on experience with LLM/agent frameworks such as LangChain, LlamaIndex, CrewAI, or AutoGen
• Demonstrated agentic AI experience: tool use, function calling, and multi-agent orchestration
• Production engineering skills: API design, containerisation (Docker/Kubernetes), CI/CD, and observability

Nice to have

• RAG expertise: vector stores, embedding models, and retrieval strategies
• Understanding of enterprise AI governance: audit trails, RBAC, and EU AI Act compliance
• Open-source AI contributions