VVZ API is not affiliated with ETH Zurich. Data might be outdated or incorrect. Please view the official ETHZ Vorlesungsverzeichnis for binding information.
AI and IT in Industry
Last Updated: 2026-06-03 00:07:32
Abstract
Participants learn how recent innovations in AI are reshaping industry, and how technologies such as large language models, recommendation systems, and reinforcement learning are transforming business operations across diverse sectors. The course combines conceptual foundations with applied insight, enabling managers to identify opportunities for AI adoption within their organisations.
Objective
Participants will understand how large language models operate and recognise their reasoning limitations; design systems that leverage LLMs for complex tasks; develop algorithms for recommendation and targeted advertising based on matrix factorisation; and model reward-based systems using Markov decision processes and reinforcement learning to build effective agents.
Content
Large Language Models (LLMs): AI systems trained on extensive text corpora that interpret, summarise, and generate human language at a high level of sophistication. LLMs are among the most readily deployable AI technologies available today, with applications spanning customer service, document analysis, and knowledge management. - Prompt Engineering: The systematic design and optimisation of instructions provided to AI systems in order to elicit accurate, relevant, and consistent outputs. For managers, this is an immediately actionable skill, as the quality of interaction with AI tools directly determines the value derived from them. - OPRO (Optimisation by Prompting): A methodology in which language models are used to iteratively refine and optimise their own prompts, reducing the manual effort involved in prompt design and improving reliability at scale. - Reasoning Models: A class of language models explicitly designed to tackle complex, multi-step problems through structured, logical processes. Understanding their capabilities and limitations enables leaders to identify where AI can genuinely augment human decision-making and where human oversight remains indispensable. - LLMs as Agents: The use of large language models as autonomous systems capable of planning, reasoning, and executing multi-step tasks with minimal human intervention. Agentic applications of LLMs have significant potential to automate complex workflows and reshape operational structures. - Recommendation Systems: AI systems that personalise content, products, or decisions for individual users by learning from historical interactions. These underpin many of the most commercially successful AI deployments to date, from e-commerce and streaming platforms to dynamic pricing and targeted advertising. - Multi-Armed Bandits: A class of algorithms that dynamically balance the exploration of new options against the exploitation of known high-performing ones. Bandits are widely used in online experimentation, advertising, and adaptive recommendation, offering a principled basis for real-time decision-making under uncertainty. - Reinforcement Learning: A learning paradigm in which agents acquire optimal behaviours by interacting with an environment and receiving feedback in the form of rewards. Reinforcement learning underlies a growing range of industrial applications, from operations optimisation and robotics to the training of modern AI systems. - Validation of AI Technologies: The systematic assessment of AI systems in terms of accuracy, robustness, safety, and fit for purpose. Rigorous validation is essential for sound procurement decisions, effective risk management, and the responsible deployment of AI in production environments.
General Information
- Language
- English
- Levels
- WBZ , NDS
- Frequency
- Semesterly recurring
Examination
- Type
- ungraded semester performance
Registration & Places
Course Components
| Type | Title | Time & Place | Hours |
|---|---|---|---|
| lecture | AI and IT in Industry | No time listed | 32 h semesterly |