Codepet journey

Building AI Products

$Code-art of floating cubes and equations in cosmic colours, formed from tiny characters$

Building AI Products

Measure Before You Optimize: The Codepet Bug That Proved Us Wrong

June 11, 20264 min read

Cinematic code-art of a glowing room of empty chairs in gold and teal characters

Building AI Products

Building an AI Product in Public: What Codepet's First Year Taught Us

June 9, 20264 min read

Code-art of a reader and a book bursting with starlight, formed from tiny characters

Second Brain

Beyond Autocomplete: Building a Second Brain That Thinks With You

June 5, 20263 min read

Code-art of a night coder at a laptop under a starry sky, formed from glowing characters

User Insights

What 500 Beginners Taught Us About Learning to Code with AI

May 28, 20263 min read

Codepet journey

How You Describe Bugs to AI Predicts Your Learning Stage

Insights in your inbox

Prompt Chaining: Break Complex AI Tasks into Reliable Steps

How to Read Any Codebase with AI: A Practical Guide

How to Reduce LLM Hallucinations in Your AI Product

How to User-Test an AI Feature Before You Ship

Designing Error States for AI Features: What to Show When the Model Fails

Model Routing: Use the Right LLM for Each Job in Your Product

How to Stress-Test Your Ideas with AI Before You Build

Why AI Feature Adoption Stalls After Week One

RAG, fine-tuning, or long context? Choose the right AI memory strategy

How to Build a Startup Knowledge Base That Gets Smarter Over Time

How to improve your AI feature after it ships

How to Cache LLM Responses and Cut Your AI Costs

How vs Why: The AI Question Pattern That Predicts Your Coding Progress

How to pair program with AI: a guide for solo builders

Your Personal Prompt Library: Stop Starting From Scratch

How to Keep a Coding Journal (and Why AI Makes It Actually Useful)

The First Session Problem: What New Users Taught Us

How to Build a Daily Coding Habit with AI That Actually Sticks

Streaming AI Responses: Why Fast Feels Smarter

The Aha Moment: When AI-Assisted Coding Finally Clicks

How to Scope an AI Feature Before You Build It

The Decision Log: Stop Relitigating Choices You've Already Made

How to Give Your AI App Memory

Spaced Repetition and AI: Make What You Learn Actually Stick

How to Write a Product Spec with AI Before You Write Any Code

How to Use AI to Turn User Feedback Into Product Decisions

What Happens When You Actually Ship: The First-Time Builder Experience

When to Use AI Agents — and When a Simple LLM Call Is Enough

Structured Outputs: The Reliability Layer Your AI Product Is Missing

How to Design AI Features That Users Actually Trust

How to Run a Weekly AI Reset (and Stop Your Project Drifting)

Comprehension Debt: The Hidden Cost of Coding With AI

LLM Tool Calling: A Practical Guide for AI Product Builders

Context Window Management for AI Product Builders

How to Give AI Better Context and Get Answers Worth Acting On

How to Debug Code with AI: What Works, What Doesn't, and Why

Prompt Versioning: Treat Your Prompts Like Production Code

How to Build an AI Learning Loop That Actually Sticks

AI as a Crutch vs. AI as a Coach: The Habit That Builds Real Skills

How to Write System Prompts That Actually Work in Production

How to Build a Personal RAG System for Your Notes

Why New Builders Hit a Wall — and How the Best Ones Push Through

How to Write LLM Evals for Your AI Product

What Is Vibe Coding, and How to Do It Without Shipping a Mess

Ask AI to Review Your Code, Not Write It

How to Use AI as a Thought Partner, Not a Search Engine

Measure Before You Optimize: The Codepet Bug That Proved Us Wrong

Building an AI Product in Public: What Codepet's First Year Taught Us

Beyond Autocomplete: Building a Second Brain That Thinks With You

What 500 Beginners Taught Us About Learning to Code with AI