Teaching

Courses and workshops

• Exploring the relationship of AI and human intelligence

  AI Exploration Week, Oxford, 6-10 July 2026 · 4 sessions · The Walter Room, Kellogg College. A history of AI from reasoning and computing to agents and the changing role of expertise.

• AI for Research(ers): Principles, Context, and Tools

  A hands-on workshop designed to move researchers from basic principles to practical application of Large Language Models. This session covers the cognitive baseline of LLMs, the nuances of context management, and the creation of custom research tools.

• Vibecoding for Language Teachers: Making apps for language learning with AI

  AULC 2026, Rewley House, Oxford · 9 January 2026. This hands-on workshop introduces language teachers to AI-assisted coding ('vibecoding') as a powerful new avenue for creating language learning applications.

Hands-on activities 20

Short activities from the Semantic Machines workshop series, each built around one idea about how language models work. They are published on semanticmachines.fyi.

• Building a Simple App

  Define the tool.

• Cognitive Baseline: Calibrating your Mental Model

  Test the model's ability to generate text and then critique its own output.

• Comparing Model Knowledge

  Test Gemini tiers.

• Context & Retrieval: Handling Information

  Test retrieval accuracy with a large document.

• Data Analysis with Code Execution

  Create a dataset.

• Deep Research: Comparing 5 Leading Tools

  Understanding the shift from conversational AI to agentic research.

• Developing a vocabulary for thinking about AI

  Foundational Concepts LLM (Large Language Model) Outputs of Neural Networks trained on extremely large amounts of data to understand and generate human-like text.

• Exploring image and multimedia generation

  The Challenge Most LLMs do not generate images directly; they call other models.

• Exploring Model Performance with MPI Explorer

  Access the MPI Explorer tool.

• Exploring Small/Local Models

  Open comparison interface.

• Literature Search Comparison

  Select a query.

• Long Context Window

  Fill the context window.

• Multimodal Conversation

  Provide visual context.

• Reflections on academic writing in the age of LLMs

  Why we need reflection To stop assuming answers are obvious and investigate gaps in knowledge regarding AI in academia.

• Search Grounding

  Test cutoff date.

• Specialized Research Tools

  Use a specialized tool.

• Testing Reasoning Capabilities

  Prepare a complex query.

• Testing the Semantic Baseline

  We will be using Google AI Studio

• Tool Use: Counting & Math

  Demonstrate the limitation.

• Visualizing Context

  Access the visualization tool.