AI Platform Engineer, Agentic Interfaces

Build and operate the interface, semantic layer and controls that let AI agents work inside IFS software safely.  Key responsibilities  Design and build MCP servers (Model Context Protocol, the emerging standard for exposing application capability to agents) over the product’s business objects, treating capability modelling, discoverability, versioning and backward compatibility as first-class design problems.  Build the write path that lets an agent safely change a customer’s operational data.  Build the semantic layer: an ontology and knowledge graph over the product, generated from what the platform already knows about itself and then curated industry by industry.  Build the skills layer that maps what someone asks for onto the correct operation and the correct sequence, with a router that picks between them.  Build the control plane: authentication, entitlements, agent identity, telemetry, metering, resistance to injection, and a default that denies rather than permits.  Build the evaluation harness that certifies agent behaviour against the real product, and improve the system against what it measures.  Build rapid prototypes and proofs of concept to validate emerging technology, product opportunities and customer scenarios.  Establish the engineering practices these systems need: evaluation, testing, observability, monitoring, governance, security and operational excellence.  Contribute to technical design, review other engineers’ work, and support colleagues coming into the domain.  Represent the work outside the team through customer engagements, demonstrations, industry events and partner collaboration.  Strong software engineering first. Everything else is applied on top of that.  Production experience building and operating enterprise systems, with real depth in distributed systems, cloud-native architectures, API and schema design, event-driven systems, security, observability and CI/CD.  Strong programming in a modern backend language. See the note on language under “Open calls” before advertising.  Experience delivering AI systems built on large language models, retrieval-augmented generation (RAG), agentic workflows and orchestration frameworks, including tool use, function calling, workflow orchestration and autonomous or multi-agent architectures, with the judgement to know where they fail.  Evaluation as a discipline: experimentation, benchmarking, prompt engineering, tracing, quality measurement and agent tuning, improving an agent against evidence rather than impression.  Ability to design solutions that integrate enterprise applications, business processes, workflows and data platforms.  Depth in at least one of the following:  Tool-surface and agent-runtime…