EC1B1 Python Coding SupportFollow-up resources for LSE BSc Economics
← Back to resources

Antigravity Resources

Contextual guide to Antigravity as a practical alternative for planning-first agent workflows in EC1B1.

Updated 2026-02-13

Why Antigravity is relevant here

In Session 4 we emphasized that tooling matters less than workflow discipline. Antigravity is useful in that context because it can support the same planning-first approach we discussed for Cursor: define the task clearly, explore options before execution, and keep review checkpoints tight.

For students choosing tools based on cost or availability, this makes Antigravity a practical option rather than a compromise.

How to approach Antigravity in this course

Use it as a structured coding assistant, not as an automatic code generator. Start from a clear brief, ask for a plan, and execute in small batches. If you keep that sequence, the quality of your outputs is usually much higher and much easier to verify.

This is especially important for coursework: code that runs is not enough. You also need coherent economic reasoning, transparent assumptions, and a clear link between your question, your data, and your method.

A simple first-week setup

Create a small project folder with one notebook and one script. Ask Antigravity to plan a narrow task, for example: data import and variable construction for one indicator. Review the plan before building anything. Then execute one step, check it, and continue.

This gives you a low-risk environment to establish good habits before moving to larger project structures.

Official links

Verification reminder

Any agent can produce confident-looking output. Treat Antigravity output as draft work that still needs your validation, especially for interpretation, causal claims, and empirical design choices.