Platform AI is not one chatbot demo. The goal is to design architecture, data boundaries, and operating rules so the application can be developed safely.
Basics
- Length: 3 days (3 × 8 hours)
- Price: from 18 000 CZK / participant / day (ex-VAT)
- Format: on-site, adapted to your Azure and data setup, on demand
- Max participants: 15
- Who it's for: AI platform teams, architects, developers, data teams, and tech leads
- Prerequisites: access to a sandbox or a clear description of the target platform
What you'll take away
- A reference architecture for a RAG or internal AI application.
- A decision model for when to use RAG, MCP, or Copilot Studio.
- Rules for data handling, security, logging, and audit.
- A prototype or technical design for one company scenario.
- A backlog of technical steps needed before production.
Program
Day 1: architecture and data
- Map the target scenario, users, and data sources.
- RAG in practice: indexing, chunking, relevance, citations, and limits.
- Security boundaries: what can be indexed and what must stay in the source system.
Day 2: tools and prototype
- Azure AI Foundry, Copilot Studio, and MCP in one architecture.
- Choosing between low-code, custom application, and controlled MCP access.
- Build a prototype or a detailed technical design for one scenario.
Day 3: operations and governance
- Monitoring, response evaluation, test sets, and incidents.
- Roles, approvals, audit trail, and sensitive-data handling.
- Plan for pilot, production rollout, and further development.
How it fits
For teams that need to prepare data first, continue with AI for Data Engineers. For security and usage rules, use AI Security & Governance.
Want this training for your team?
Every training is tailored to your team. Tell us about your people, stack, and goals, and we'll come back with a concrete proposal.
Want to stay one step ahead?
Don't miss our best insights. No spam, just practical analyses, invitations to exclusive events, and podcast summaries delivered straight to your inbox.