← Back to training

Trainings

From AI chat to agents: automate your workflows with Claude Cowork

A one-day training that takes you from basic AI chat use to leveraging agentic tools, showing how to automate routine knowledge work in practice. We cover what language models and agents actually are, where they deliver genuine value and where they should not be used. During the day we automate a concrete workflow together and build a foundation you can carry forward independently. By the end you can hand repetitive routines like research or reporting to an agent safely and reliably. No coding skills required.

Duration
One day, 9am–4pm
Format
On-site at your office or remote
Language
Finnish or English
Group size
We recommend up to 20 participants so everyone gets personal support
Price
EUR 2,500 (+ VAT), Tampere and Helsinki metropolitan area. Ask for a quote for other locations.

Antti used to be really AI-sceptical, and now after the training he is super excited. And by golly, he had built a slick tool with Cowork that he showed off yesterday.

A participant's manager

Who is this for?

This training is for knowledge workers who spend their days with documents, spreadsheets, email and reports, and who want to automate repetitive routines with agents. You get the most out of it when you already use AI occasionally and want to move from one-off prompts to repeatable workflows. No coding background needed.

  • Sales and marketing: prospecting, content, account research
  • Finance and administration: reporting, data consolidation, routine calculations
  • Specialists and managers doing a lot of repetitive research work
  • Teams that want to integrate AI into their everyday ways of working

Agenda

The day is built around four blocks: the morning builds a solid foundation in language models and ground rules, then advances to agents, skills and connectors. In the afternoon we do hands-on exercises with Claude Cowork: picking a routine task and automating it.

09:00–10:30: Language models and key concepts

  1. What a language model is and how it actually works: statistical prediction, not a thinking machine
  2. The pitfalls of language models: hallucinations, sycophancy and bias, and how to guard against them
  3. Data privacy and security: what you can and cannot feed an AI
  4. The EU AI Act in practice: what is prohibited and what counts as high-risk use
  5. Which model for which job? Balancing intelligence, speed and cost

10:45–12:00: Agents, tools and connectors

  1. From chat to action: how AI chat, retrieval (RAG) and an active agent differ
  2. The agentic loop: plan, act, check, correct
  3. Agent vs. traditional automation: when to use which
  4. Skills and subagents: teach your agent repetitive tasks once and for all
  5. MCP and connectors: how the agent connects safely to email, calendar and company data

13:00–14:15: Hands-on with Claude Cowork

  1. Your first working workflow in Claude Cowork from start to finish
  2. Prospecting in practice: the agent builds a target-company list with background research
  3. Financial reporting in practice: turning numbers into a plain-language summary for management
  4. Consolidating data from PDF files and spreadsheets into a single view
  5. Context management: ensuring the agent has the right context for reliable, accurate results

14:30–16:00: Build your own skill

  1. Identify a task in your own work that you keep re-explaining to AI
  2. Write, test and refine your own skill, ready to share with your team
  3. Connect your first connector safely and verify its operation
  4. Build your own or use off-the-shelf: the benefits and risks of ready-made skills
  5. How to continue tomorrow: a concrete plan to streamline your everyday work

By day's end you'll know how to:

  • Understand and explain what language models and agents are and what they can do
  • Spot the tasks in your own work that are worth automating
  • Run a real workflow in Claude Cowork from start to finish
  • Build, test and deploy your own routine tool (skill)
  • Connect a connector and give the agent safe access to needed data
  • Use AI responsibly and safely: data privacy, bias and regulations under control

Instructor

Veli-Pekka Eloranta

Veli-Pekka Eloranta

Curious Trainer

Since 2003 I've been doing IT in every role imaginable: coding, coaching and leading teams. Today my focus is clear: I help organisations transform and harness AI agents as part of agile ways of working.

wellu@curious.fi
044 557 0591
LinkedIn →

← Back to training