Agentic AI: Beyond Coding Assistants and Chatbots

A hands-on workshop on building AI agents step-by-step.

Learning goals

This is a hands-on workshop on building AI agents that can act, not just answer. Over the sessions, participants build agents step by step, starting from a plain OpenAI REST API call and ending with complex multi-agent workflows using Pydantic AI. Along the way the workshop covers the full stack you need to ship agents: the components that make up an agent, how they connect, and how to run them for specific task with a limited scope.

The topics covered include:

• LLM basics and inference: how models work at a practical level, serving models with an inference server like Aitta, and using prompt templates.
• Building agents from scratch: from a single REST call to structured agentic workflows with Pydantic AI.
• Components of an agent: the model, the harness, the loop, tools, MCP servers, skills, memory, and how they fit together.
• Security, evaluation: what breaks when agents leave the laptop, how to evaluate them, and how to run them reliably.

Key takeaways for participants:

• Identify the main components of an AI agent (model, tools, memory, harness) and recall standard concepts such as tool calling and MCP.
• Build working agents in stages, from a raw OpenAI REST API call to structured agents with Pydantic AI, and connect tools, MCP servers and memory to them.
• Analyze and identify potential pitfalls when moving agents from development to production setting.

Target audience

• IT professionals, Software engineers and data scientists
• Programmers curious to move out of a chat interface and integrate AI in their routine tasks.
• People who have used LLM chatbots or coding assistants and want to understand how they work behind the scenes.

Prerequisites

• Prior programming experience in Python.
• No prior machine learning background is required

Agenda

2026-11-02: Pre-workshop setup with Aitta and the training environment
2026-11-03: LLMs and agents with simple tool calling.
2026-11-04: Agentic workflows with advanced capabilities such as MCPs, persistence, memory and skills.
2026-11-05: Evaluation, risks and security of agentic workflows

Schedule

Setup and dry-run:
2026-11-02 (Mon)
Sweden: 13:00 – 14:00
Finland: 14:00 – 15:00

Main event:
2026-11-03 (Tue) to 2026-11-05 (Thu) – Afternoons 3 hours each day
Sweden: 12:30 – 15:30
Finland: 13:30 – 16:30

The workshop is Online

The LUMI AI Factory Service Center is funded jointly by the EuroHPC Joint Undertaking and the participating states FI, CZ, DK, EE, NO, PL. Sweden AI Factory is hosted by NAISS (the National Academic Infrastructure for Supercomputing in Sweden) in partnership with RISE Research Institutes of Sweden. Funding is shared between Sweden (through the Swedish Research Council and Vinnova, Sweden’s Innovation Agency) and the EuroHPC Joint Undertaking, which is the EU’s supercomputer organisation.