[Masterclass for Pro] Part I: Modern gAI Skills for Scientists

Instructor:

Marin Bogdan, Msc Phyics ETH Zürich

Independent IT consultant specialized in AI, project management and information security
Quality manager and AI lead for Swiss E-Voting, Swiss Post Ltd.

Topics Covered

The AI Ecosystem

Types of AI Tools

This section addresses how to use different categories of AI tools optimally in scientific work. Generic assistants and specialized tools serve different purposes, and understanding which to reach for — and when — makes a measurable difference in both productivity and the quality of your output.

The Big AI Players & Pricing

An overview of the main AI providers in each tool category, along with practical pricing information, so you can make informed decisions about which subscriptions are actually worth adopting for your own work.

Beyond Prompting: Modern Techniques for Using AI Efficiently

AI Workflows in Science: Overview of Claude

This segment surveys the full range of tools within Anthropic’s Claude ecosystem — Claude.ai, Claude Desktop, Claude Cowork, Claude Code, Claude extensions for IDEs, browsers, and office apps, Claude Science, and the Claude API — and how each fits into different scientific workflows. Claude serves as the illustrative case throughout the training, though equivalent tools from other vendors are discussed where relevant.

AI as a Critical Co-thinker, Not Just a Helpful Assistant

This topic moves beyond simple chatbot prompting into context engineering: structuring your interactions with AI to get more reliable results, reduce hallucination, and avoid “context rot” (the degradation of output quality as conversations grow long or unfocused). The goal is to shift the AI’s role from passive responder to active, critical thinking partner.

Data Protection and Legal Requirements

A practical look at how to protect your data and meet legal and institutional requirements when using AI tools — an essential consideration for anyone handling research data, student information, or sensitive materials.

Applied AI Workflows for Your Own Projects

Working with Documents

Hands-on use of Claude Cowork and Claude Code to work directly on your own documents and projects, moving from theory into applied, project-specific practice.

Workflow Automation & Skills

Learn to extend your AI with custom skills so it can apply your expertise and carry out complete workflows on your behalf, rather than responding to isolated prompts. This section introduces Skills, Plugins, Agents, MCP Servers, Automation, Remote Work, Context Management, and Markdown — the same building blocks explored in greater depth in Part II.

Research

Practical use of AI in scientific research: searching the literature, verifying claims, and auditing outputs for accuracy and reliability.

Technical Requirements and recomendations

Attendees must arrive with the following set up in advance:

• A laptop/notebook computer

• Free subscriptions to: OpenAI ChatGPT, Anthropic Claude.ai, and Google Gemini (Mistral optional)

• Desktop versions of ChatGPT and Claude installed

• Strongly recommended: a Claude Pro subscription (EUR 20/month) or Max subscription for Claude Code

Note: Several requirements or recommendations involve paid subscriptions and multiple software installations. Attendees should anticipate setup time beforehand to avoid delays on the training day. Participation is also possible without a subscription, although hands-on personal practice may be limited. While the workshop uses Claude/Anthropic as the primary illustrative example, it also discusses equivalent alternatives from other vendors.