32dots HEIDELBERG AI
EXTENDED COURSE

Cursor

Cursor is VS Code reimagined as an AI code editor: inline autocomplete, a chat sidebar that understands your own files, and an Agent mode that writes, runs, and fixes multi-file changes from a plain-English request. This hands-on course takes you from installing Cursor to shipping a real feature with the agent — then shows the pro moves: picking the right model per task, feeding the agent the right context, and keeping your costs under control (including the free year of Pro for students).

1Lessons7step-by-step, ~90 min each
2Cheat sheetcopy-ready expressions
3Examples3what people built

Dashboards & analytics1

Cursor AI-native SME Founder
In the gallery

Find and optimise slow SQL queries in an existing ORM layer

Cursor reads the SQLAlchemy model definitions and the slow-query log, identifies N+1 patterns and missing indexes, rewrites the problematic queries using eager loading, and generates the Alembic migration to add the indexes.

Dashboard load time drops from 4.2 s to 0.6 s; the migration is production-ready with rollback included.

Try it yourself

Open Cursor chat with the ORM files in context: "Here is the slow-query log. Find N+1 patterns and missing indexes in models/. Rewrite the queries and generate an Alembic migration for the recommended indexes."

Internal tools & ops1

Cursor Founder Creator
Course starter

Refactor a monolithic React page into composable components

Using Cursor's repo-wide context, a 600-line React page component is broken into eight focused sub-components, props are typed with TypeScript interfaces, and shared state is moved into a context provider — without changing any visible behaviour.

The page becomes composable and testable; two new features are added the same afternoon by reusing the extracted components.

Try it yourself

Open the file in Cursor, press Cmd+K: "Break this page into composable sub-components. Type every prop with TypeScript interfaces. Extract shared state into a context. Do not change any visible UI behaviour."

Forms, surveys & feedback1

Cursor Scientist Creator
Course starter

Add interactive charts to a survey results page

Cursor reads the existing survey data schema and a static results page, then adds Recharts bar and Likert-scale charts that pull live data from the API — matching the existing Tailwind design system and responsive layout.

Survey results are visual and interactive instead of a wall of numbers; stakeholders explore the data directly without needing an export.

Try it yourself

In Cursor chat: "Add Recharts bar charts and a Likert-scale heatmap to pages/results.tsx. Pull data from the existing /api/survey-results endpoint. Match the Tailwind classes already used on the page."