Senior AI Engineer (Netherlands)

ph3About the Role: /h3 pWe’re building AI into STARLIMS, a platform used across quality manufacturing, life sciences, public health, forensics, and environmental sciences. /p h3About the Role: /h3 pWe’re building AI into STARLIMS, a platform used across quality manufacturing, life sciences, public health, forensics, and environmental sciences. /p pThis role is focused on agentic systems: software that reasons over a task, calls tools, works through multiple steps, and hands the result to a person to review and approve. /p pOur users work under strict accuracy, traceability, and validation requirements. The engineering challenge is making non-deterministic systems reliable, observable, and controllable enough to be trusted, tested, and shipped. /p pYou’ll work on both the platform and runtime our agents execute on and the production agents built on top of it. /p h3What You’ll Work On: /h3 h3Agent Platform Runtime (Core Focus) /h3 ul liDesign and build the runtime our agents execute on: planning and execution loops, tool calling, state management, durable execution, and failure recovery /li liBuild the layer through which agents reach platform data and external systems safely /li liDesign coordination, delegation, and handoff across agents and workflows where needed /li liMake agent behavior versionable, testable, measurable, and regression-safe across releases /li liBuild reusable primitives so new agents are configured rather than rebuilt from scratch /li /ul h3Building Agents (Core Focus) /h3 ul liTake a domain workflow from expert conversation to a working agent: goals, actions, execution flow, failure handling, and success criteria /li liGround agent decisions and outputs in authoritative enterprise data rather than relying on model knowledge alone /li liImplement human-in-the-loop by design, including approval gates, override capture, uncertainty handling, and clear evidence for agent decisions. Agents recommend and draft; people decide /li liClose the loop: turn user corrections and overrides into signals that measurably improve the agent /li /ul h3Evaluation Reliability /h3 ul liBuild evaluation harnesses for multi-step behavior, not single-response accuracy: task completion, tool-call correctness, groundedness, trajectory quality, and regression across model, prompt, and tool changes /li liDefine production metrics for agent quality, reliability, latency, cost, and human intervention rates /li liImplement guardrails, fallbacks, timeouts, cost ceilings, and end-to-end observability and tracing across agent runs /li liDesign safeguards against prompt injection, unsafe tool use, excessive permissions, data leakage, and other agent-specific security risks /li liManage prompt evolution, model drift, and non-determinism while maintaining consistent, measurable system behavior across releases /li /ul h3Integration Data /h3 ul liIntegrate agents with platform APIs and third-party enterprise systems already running in our customers’ environments /li liBuild retrieval and context pipelines that turn fragmented enterprise data into reliable, permission-aware agent context /li liDesign controlled execution paths for automated actions, with a complete, traceable audit trail /li /ul h3Platform Infrastructure /h3 ul liBuild and operate backend services on AWS (Lambda, API Gateway, DynamoDB, Step Functions, etc.) /li liOwn significant parts of the system architecture and contribute to key technical decisions /li liContribute to infrastructure-as-code and deployment pipelines /li /ul h3Tech Stack /h3 ul liLanguages: TypeScript, Python /li liBackend: Node.js, Python, AWS Lambda, Step Functions /li liAI: OpenAI, Anthropic, MCP and related agent/tool protocols, embeddings and vector search /li liFrontend: React, Next.js, Tailwind CSS /li liInfrastructure: AWS, Terraform /li liTesting: Jest, Playwright, pytest /li /ul h3What We’re Looking For /h3 h3Must Have /h3 ul li6+ years of software engineering experience, including production systems /li liExperience building production LLM systems, including tool-using or multi-step agentic workflows beyond simple prompting and chat interfaces /li liStrong understanding of LLM behavior, limitations, and failure modes, especially how errors compound across a multi-step run /li liExperience with LLM APIs, tool and function calling, and designing planning and execution loops /li liExperience evaluating and debugging non-deterministic systems /li liSolid backend and cloud experience (AWS or equivalent) /li liProficiency in TypeScript and/or Python /li /ul h3You Should Be Comfortable With /h3 ul liDebugging across distributed and non-deterministic systems /li liMaking explicit tradeoffs between accuracy, latency, reliability, and cost /li liWorking in ambiguous problem spaces where the right architecture isn't obvious yet /li liOwning production systems end-to-end /li liChoosing conventional software over AI when AI isn't the right solution /li /ul h3Nice to Have /h3 ul liC#, Microsoft .NET Framework /li liTool and interop protocols such as MCP /li liEvaluation pipelines and metrics built specifically for agentic systems /li liExperience in regulated or domain-heavy systems (validation, audit trails, controlled change) /li liRetrieval and grounding techniques for supplying agent context /li liWorkflow and durable-execution platforms (Temporal, Step Functions, n8n, etc.) /li liContainerization and orchestration (ECS, EKS, Kubernetes) /li liInfrastructure as Code (Terraform or similar) /li /ul pSTARLIMS is an Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, creed, religion, color, national or ethnic origin, citizenship, sex, sexual orientation, gender identity and expression, genetic information, veteran status, age or disability status. /p /p #J-18808-Ljbffr