Research & Scientific Computing with AI. Use AI across your research, and use it well.

Where AI is making a difference in research today, and the ethics and norms that govern its use.

Topics covered

• Where AI is being used across the research cycle
• Research norms: what journals and supervisors expect when you use AI
• Choosing where AI fits in your own research

You produce: a map of AI uses and norms in your own field.

Find, read and organise papers with Perplexity, Semantic Scholar and NotebookLM, without losing citation discipline.

Topics covered

• Searching for papers with Semantic Scholar and Perplexity
• Summarising and questioning papers with NotebookLM
• Managing references with BibTeX
• Citation discipline: checking every source an AI suggests

You produce: an annotated literature map for your topic.

Use AI to draft, edit and structure scientific writing while keeping your own voice and rigour.

Topics covered

• Structuring papers, reports and thesis chapters
• Drafting and editing with ChatGPT and Claude
• Writing in LaTeX with BibTeX references
• Keeping your voice and rigour in AI-assisted text

You produce: a drafted and edited paper section.

Take Python analysis beyond Foundation level into the workflows of your own domain.

Topics covered

• Going beyond Foundation-level Python and NumPy
• Scientific computing with SciPy: fitting, optimisation and statistics
• Building domain-specific analysis workflows
• Structuring larger analyses across notebooks and scripts

You produce: an analysis notebook built on a workflow from your field.

When a simulation is the right tool, how to build a simple one, and which tools each domain uses.

Topics covered

• When to simulate, and when to measure or calculate
• Building a simple model in Python with NumPy and SciPy
• Simulation tools used in different domains
• Checking a model against known results

You produce: a simple simulation of a system from your field.

Make publication-quality figures that meet what reviewers expect to see.

Topics covered

• Choosing the right figure for your data
• Building publication-quality figures in Python
• What reviewers expect: labels, error bars, colour and resolution

You produce: a set of publication-ready figures.

Make your research reproducible with Jupyter, Git, managed environments and versioned data.

Topics covered

• Organising a research project in Jupyter
• Version control with Git
• Managing Python environments and dependencies
• Versioning your data alongside your code

You produce: a reproducible research repository.

Write, debug and optimise research code with AI assistants.

Topics covered

• Writing research code with ChatGPT and Claude
• Debugging errors with AI help
• Optimising slow code, and checking the results still hold
• Testing AI-written code before you trust it

You produce: a debugged and optimised research script.

Use AI responsibly: attribution, hallucination risks and the norms of peer review.

Topics covered

• Attribution: declaring how you used AI
• Hallucination risks: invented results and references
• Peer review norms and the use of AI

You produce: an AI-use statement for your own work.

Produce a research artefact using AI throughout, and document your AI workflow.

Topics covered

• Choosing an artefact: a mini-paper, a literature review or a thesis chapter draft
• Using AI at each stage, from reading to writing
• Documenting your AI workflow

You produce: your research artefact and AI workflow log.