Using AI tools in academia (4h)
Types of generative artificial intelligence systems used in professional activity and choosing the right tools for specific tasks. Comparative presentation of AI tools for text generation and processing (ChatGPT, Claude, Gemini). Techniques for effective formulation of requests (prompt engineering). The difference between generic question and structured request. Characteristics of an effective request, RCTC model: role, context, task, constraints. Verification, validation and review of AI-generated content: identification of errors, reduction of the risk of inaccurate information, formulation of requests for clarification. Identification of hallucinations, cross- checking techniques and validation of sources. Ethical and academic integrity norms for AI use.
Practical application: Using specialized AI tools to summarize key concepts and create reports on a specific topic including checking bibliographic references
Using AI tools for productivity (Microsoft Copilot) (4h)
Introducing Microsoft Copilot as an AI-based teaching and research productivity tool, integrated into Microsoft applications (Bing, Outlook, Teams, etc.) that use large linguistic models (LLM).
Presentation, example and discussion of AI functions for productivity applications in the Microsoft suite:
- Text editing and summarization
- Data analysis, processing and visualization
- Create and edit presentations
- Summarizing, retrieving information and making decisions in meetings
Automate repetitive tasks through Copilot software agents.
Practical Application: Using Microsoft Copilot Notebooks/Google Gemini Notebooks for task organization and productivity
Applications of data science in engineering (4h)
open science datasets . Data science methodology from an interdisciplinary perspective: computer science + statistics + domain expertise. Data types and data quality indicators. Pre-processing methods for technical datasets. Exploratory data analysis (EDA) and advanced visualization methods. Predictive data modeling and anomaly detection methods. Open-source software libraries for data science e.g. Python pandas , sci-kit learn , pytorch , matplotlib/se aborn. Using AI tools to automate workflows in engineering data analysis and processing.
Practical Application: Using IEEE Dataport/Zenodo for searching, analyzing, and preliminary evaluation of engineering datasets