Jobid=631039727021525294 (0.0992)
Job Description
Build the intelligence behind the experience: the session and memory model, retrieval across our own estate and the documents that run it, the agentic flows that carry out multi-step work, and the evaluation that tells us whether any of it is right.
What you will do
-
Build the session and context spine: persistence, resumption, hand-over with the reasoning trail intact, and context resolved from a person's identity and entitlements.
-
Build retrieval and grounding across structured records and documents, at a cost and latency the product can afford.
-
Build document intelligence over manuals, engineering drawings, certification directives and contracts, extracted and citable, because for many of our customers those carry as much operational truth as the databases.
-
Build the agentic flow runtime: multi-step work with tools, and the rules that decide when a human is asked rather than told.
-
Handle speech and images as first-class input alongside text, so the same request works dictated in a plant room or photographed at an asset.
-
Build the evaluation harness, and treat it as the gate on every claim we make about autonomy.
-
Build the language interface over our scheduling and optimisation engines, expressing trade-offs in the customer's own terms.
-
Build the security layer with our security team: resistance to prompt injection, personal data handling, guardrails, audit and data residency.
Qualifications
- Experience with S2S integration, preferably using technologies such as n8n or Temporal, or alternatively MuleSoft, Apache Camel, or Boomi.
- Experience with Go and Python.
- Experience building multi-tenant SaaS solutions.
- Experience building and scaling AI-native products and applications.
- Experience with production-grade GenAI solutions, including RAG pipelines (hybrid retrieval, embeddings) and agentic systems (agent orchestration, tool usage).
- Experience with AI frameworks and tooling such as Pydantic AI, LangChain, and MCP.
- Experience with MCP development and plan-based agentic software development.
- Experience with DevOps and cloud-native infrastructure, including Docker, Kubernetes, CNCF technologies, and CI/CD pipelines.
- Experience with API gateways and API products, preferably with platforms such as Kong or Tyk.
- Preferably experience optimizing AI services for latency and cost.
- Preferably experience with specification-driven development.
- Passionate about and eager to adopt new technologies.
- Up to date with the latest developments in GenAI and AI-based software development.
Additional Information
Deel deze vacature: