Pair Programming Habits with an AI Assistant
Work with a coding assistant like a good pair: small steps, clear roles and code you understand
A taste of a lesson
I asked the assistant to build my whole login page and got 300 lines. It mostly works but I don't understand half of it. What should I do?
Pause before building on it. Code you do not understand is hard to debug later. Two options. Ask the assistant to walk you through it section by section, and question anything unclear, especially password handling and session code. Or set it aside and rebuild in small steps: form first, then validation, then the request to the server, running each piece. Either way, commit only what you can explain. Try this: pick the part you understand least and ask the assistant why it was written that way.
Written by the teacher as an example. In your lesson the tutor answers your own questions, and like any AI it can be wrong.
What you will be able to do
- Choose between completion, chat and agent help for different tasks
- Work in small, tested steps and commit working states often
- Review and question every suggestion before accepting it
- Recognise when to stop prompting and think or restart with a clean summary
Lesson plan
- 1 Three ways an assistant helps Match completion, chat and agent modes to the right kinds of work. Start
- 2 Roles: driver and navigator Keep yourself in charge of decisions while the assistant helps. Start
- 3 Small steps and short loops Work in increments you can run, test and understand. Start
- 4 Reading before accepting Review suggestions critically and ask good follow up questions. Start
- 5 When to stop prompting Recognise loops and dead ends and recover from them. Start
- 6 Privacy and team rules Use assistants within your employer's and clients' rules. Start
Try asking
About this tutor
For developers starting to use AI coding assistants in their editor or chat, and for those who already do but feel they are losing track of their own code. You will build working habits borrowed from human pair programming: deciding who drives, keeping changes small, thinking aloud with the assistant, asking for explanations before accepting code, running and testing after each step, and knowing when to stop prompting and think. Each lesson includes a short exercise in your own project and language. The aim is faster work that you still fully understand.
Reviews
4.3
3 ratingsSample
- Nina P.Sample
Small steps and committing often sounds basic, but it changed how I work. I no longer end up with huge AI changes I am scared to touch.
- Freya D.Sample
Good habits for beginners. Experienced devs might find the first lesson slow, but the rest is useful.
- Kareem A.Sample
The 'stop prompting after two failed fixes' rule saves me so much time. Friendly and practical.
About the teacher
I teach developers and beginners the daily habits that make AI coding assistants useful instead of risky
9 tutors 411 lessons taught Sample
I am a software developer who teaches the working habits around AI coding assistants: how to describe a task, how to pair with an assistant without losing the thread, how to use a terminal based coding agent, and how to write project instruction files that keep it on track. I teach from the keyboard. We open a small project, try...
See Jude's profile and tutorsMore like this
Other tutors on the same or nearby topics.