Description
Artificial intelligence has the potential to fundamentally change work in laboratories, analytics, quality management and R&D.
Since the introduction of ChatGPT 2022, many have gained initial experience with so-called Large Language Models (LLMs). Although these models are not the same as "AI", they form the language-understanding backbone of modern AI systems. They write texts, analyze content, generate program code, provide feedback and even help to solve complex tasks - if you know how to work with them.
This is exactly where prompting comes in: If you ask the right questions and give precise instructions, you will not only get better results, but also unlock the full potential of these tools - as creative idea generators, analytical assistants or virtual consultants.In addition to sound know-how, you will receive directly applicable tips and many suggestions for use in everyday laboratory work.
The seminar is part of Klinkner & Partner's "AI in the lab" series and is aimed at anyone who wants to use AI tools professionally in a scientific and technical environment.
The seminar is aimed at:
- Specialists and managers from laboratories, analytics, QM and QA
- Employees in research & development
- AI users with little knowledge of AI
- Anyone who wants to make their work more efficient with AI
After attending the seminar
- have an overview of the current market and the areas of application of LLMs (ChatGPT, Gemini & Co.)
- know how to formulate prompts clearly, effectively and confidently
- you can use instructions, context and role correctly for prompts
- understand how AI can be used for research, evaluation and reporting and how routine activities can be automated.
Program
Overview & tools in comparison (Freichel)
- Classification of LLMs in the AI tool landscape
- Basics of LLM terms: what are tokens, parameters, context windows...
- Overview: ChatGPT, Gemini, Perplexity, Claude & Co. in direct comparison
- Differences in database, architecture, source citation and additional functions
- Data protection, cost models and legal aspects of AI use
- Risks when using AI: distortions, hallucinations and bias
- Live demo: Same request - different tools - surprising differences!
Understanding & applying prompting (Freichel, Harringer)
- Anatomy of a good prompt: role, task, goal, format
- Improve output through iteration and feedback loops
- Assessing AI results correctly: Quality, sources, control
- Dealing with boundaries, ambiguities and critical issues
- Mini-exercises directly in the tool of your choice - easy to implement even for beginners
Typical use cases from laboratory, analytics & quality management (Hunke, Harringer)
- AI-supported optimization of your email communication
- Specialist research: standards, procedures, device manufacturers, specialist literature
- Reports & findings: test reports, calibration certificates, summarize test results in a structured manner
- Writing, checking and improving QM documents, more productive audits with AI, documenting findings
- Text optimization: spelling, style, comprehensibility, length, target group adaptation
- Tips for handling sensitive and confidential data
- In small groups: Work on your own use cases, develop prompts together and test them directly
Automating with prompts: data, routines & agents (Hunke)
- Save time with prompts: process recurring tasks efficiently
- Result calculation & pattern recognition: Evaluate measurement data (e.g. Excel data, CSV tables, measurement series)
- Typical cases: Concentration determination, calibration curves, trends in stability tests
- AI as a routine helper: generating logs or draft reports, summarizing data, documenting to-dos
- Current, practical applications for intelligent AI support without programming knowledge
- Outlook for the future: AI-supported assistants and agents in the day-to-day work of laboratories
- With practical examples and live demonstrations from the laboratory world - plus an outlook on automation potentials
You can find the complete program here.
