🔍 Advertising Regulation in Japan (Promotional Material Review) and Artificial Intelligence JP/EN
Ethics · Regulation · Technology — Pharma Practice Notes
Home  /  AI Programming
Educational·10 episodes total

AI Programming── Code written with AI, code written by AI — interrogating the boundary in 10 episodes

The work of writing code is in deep transition. Do humans write and AI reviews, or does AI write and humans review? In an age where one line of prompt produces hundreds of lines of code, what should be called "the programmer's skill"? This series covers, in 10 episodes, everything from practical developer work to the big questions.

Introduction
01

Intro to AI Pair Programming

GitHub Copilot, Claude Code, Cursor ── the differences between the three major tools, and when to use which. The first step toward introducing AI "as a partner" to real work.
Basics
In preparation
02

Prompt Design for Code Generation

The gap between good and bad instructions changes output quality by 10×. Patterns and anti-patterns for prompting for code.
Basics
In preparation
03

AI-Assisted Debugging

Paste error messages to AI and reverse-engineer the root cause. The full flow: reproducing, hypothesizing, validating fixes.
Basics
In preparation
Daily work
04

Querying Databases in Natural Language

"Who were the top 10 users last month?" — the era where you write in English (or Japanese) and AI generates the SQL. Where Text-to-SQL fits in practice.
Work
In preparation
05

Reading Legacy Code

Use AI to "translate" 10-year-old code nobody wants to touch. Inventorying technical debt and prioritizing refactoring.
Work
In preparation
06

Automated Test Generation

Let AI think up edge cases. Quality assurance in an era when AI proposes "ways to break it that humans wouldn't imagine."
Work
In preparation
07

AI Code Review

Have AI sniff out code "smells." Splitting reviewer effort between mechanical checks and semantic discussion.
Work
In preparation
Architecture and beyond
08

Architectural Design as a Dialogue

Microservices? Monolith? Event-driven? Discuss the large structure with AI to organize the trade-offs of each option.
Future
In preparation
09

No-code / Low-code × AI

The era when non-engineers can build. New roles and new problems arising from non-engineer × AI collaboration.
Future
In preparation
10

Software Engineering in the Age of AI

Tests, CI/CD, design reviews, requirements ── what remains, what disappears. The "next shape" of the programmer profession.
Future
In preparation

— Each episode will be published on this page once it has passed verification —

← Back to home