
Integration Objects
TunisiePublication : 20 nov. 2026
Development - AI Copilot for Natural-Language Industrial IoT Platform Configuration via MCP
- Stage
- 4-6 mois
- Présentiel
- Échéance : 4 déc. 2026
- MCP
- LLM/RAG
- LLMs, Generative AI, RAG & AI Agents
- LLMs, Generative AI, RAG & AI Agents
- MQTT and IIoT
- SIOTH UHMI
- .NET/C#
Description du poste
Abstract: Build an MCP (Model Context Protocol) server that exposes SIOTH® configuration and runtime APIs as secure tools, and an AI assistant enabling engineers to configure, query and troubleshoot SIOTH in natural language.
Main items:
- Study MCP, LLM tool calling and SIOTH APIs
- Design MCP tools, resources and access rights
- Develop the SIOTH MCP server in .NET
- Build the assistant with preview and human approval
- Evaluate accuracy, security and response time
Assistant capabilities:
- Create/configure connectors, tags and data flows
- Set up historian archiving, rules and workflows
- Query live values, connector status and logs
- Troubleshooting guidance with diagnostics and docs
- Bulk configuration from files (CSV, Excel, tag lists)
Deliverables: SIOTH MCP server source code and tool catalog; configuration assistant with UI; documentation (report, design, test reports).
Expected results: Faster SIOTH project setup; lower learning curve; safe and auditable AI-driven configuration.
Required application document: CV.
Keywords: MCP, LLM, Generative AI, AI Agents, IIoT, SIOTH, .NET/C#.