AI 4 Coding

Day 1: Course Overview

Eduard Bukin

ebukin@worldbank.org

Distributional Impact of Policies. Fiscal Policy and Growth Department

2026-04-20

Motivation

Chat-Based AI tools Claude | ChatGPT | Gemini | MAI

• Use familiar technologies: 🌐 web browser, 💬 chat, 📝 Stata editor, 📋 copy-paste.
• Improve our workflow.

Is this the most effective way to use AI in coding❓

There are many Integrated Development Environments (IDEs) for data science with AI as a part of your coding workflow. Introducing Positron with GitHub Copilot is the goal of this course.

See also: previous seminar on AI-Assisted Coding in Stata and R on GeoPovlearn.

Course Overview

Goal: Equip economists with AI-skills for coding.

2 × 3-hour sessions Self-study in between GitHub Copilot · Positron IDE Stata & modern stat languages

Understand

• Core LLM concepts: tokens, prompts, context windows, agents
• How these affect AI-assisted coding
• AI’s strengths and limitations

Practice

• Use AI to write code and reuse old code
• Plan execution and write prompts for multi-step tasks
• Supervise and course-correct AI-generated code
• Leverage out-of-the-box tech: #tools, skills, /agents, /prompts, etc.

Take Control

• Adopt decent workflows for AI-assisted coding.
• Comply with data-safeguard practices.
• Use AI responsibly for coding.

Course content

Day 1

Basics

• Software setup & walk-through
• Key concepts: IDE, assistants, context, tokens, LLM, and how it works
• Examples: data preparation and cleaning, and security techniques
• Common techniques: managing AI’s context and securing your data

Self-Study

The most important part

• Practice AI use with your own code
• Structured examples and learning materials to guide you
• Examples:
  • Testing your setup
  • Explore positron (self-study)
  • Understanding data with AI (live)
  • Reproduce, Revise, and Reuse existing analysis
• Come back with questions & observations

Day 2

Go deeper

• Debrief self-study experiences
• Examples: planning, using #tools, skills, /agents, and /prompt-files.
• Other relevant techniques

Logistics

	Morning	Evening (repeat)
Format	In person + virtual	Virtual only
Day 1	Mon Apr 20, 9 AM–12 PM ET	Mon Apr 20, 9 PM–12 AM ET
Day 2	Thu Apr 23, 9 AM–12 PM ET	Thu Apr 23, 9 PM–12 AM ET

• Evening repeat session is for those who cannot attend the morning session. It will cover the same material, but it will be virtual only.

• Questions during the session: use the Teams chat or raise your hand.

Today’s Agenda

Time	Topic
9:00-9:10	1. Welcome and course overview (slides)
9:10-9:40	2. Setup overview (slides). Example 0 (live) and Example 1 (self-study)
9:40-10:20	3. Example 2: Understanding data with AI (live)
10:20-10:30	☕ Break (10 min)
10:30-10:50	4. Core AI-related concepts
10:50-11:00	5. Common workflows and data safeguarding (slides)
11:00-11:30	6. Example 3: Reproduce, Revise, and Reuse existing analysis (live/self-study)

Course resources

worldbank.github.io/ai4coding

• References, detailed explanations, guided self-study exercises and recordings, along with in-class materials.

github.com/worldbank/ai4coding-examples

• Code for exercises and examples.
• Also available on OneDrive for WB staff: ai4coding-examples

ai4coding-data - data for exercises (OneDrive - WB only)

Previous seminar on AI-Assisted Coding in Stata and R on GeoPovlearn

What next?

1. Set up the software.

2. Set up your course working folder and copy relevant materials into it.

3. Follow along with the materials for Day 1.