How to use Claude Code like the people who built it
The founding engineers' own prompting and planning habits.
Open every.to →
Building & Product
Podcast
Podcast
Guide
Article
Article
Guide
Guide
Guide
Guide
Article
Guide
Article
Guide
Article
Guide
Article
Article
Tool
How do I plan and prompt so the AI builds what I actually want?
The single biggest upgrade is to stop prompting features and start with a spec: brainstorm the idea into a written plan (a mini PRD), break it into small steps, then feed the AI one step at a time. Harper Reed's spec-then-plan-then-execute loop and Lovable's Prompting Bible are the two canonical playbooks. Small, specific, one-thing-at-a-time prompts with real context beat long wishlists every time, and they burn far fewer credits.
Go deeper
18 resources.
How two engineers ship like a team of 15 with AI agents
A real company's compounding-engineering workflow, shown live.
Open every.to →My LLM codegen workflow atm
The most-copied spec-first workflow, with the exact prompts.
Open harper.blog →An LLM codegen hero's journey
How the workflow evolved after months of real use.
Open harper.blog →Simon Willison on Harper Reed's workflow
A trusted second opinion on why the approach works.
Open simonwillison.net →The Lovable Prompting Bible
The official handbook: four levels of prompting, from training wheels to meta.
Open lovable.dev →Prompting best practices (Lovable docs)
The concise, always-current version of the Bible.
Open docs.lovable.dev →Vibe coding: best practices for prompting
Backend-aware prompting patterns from the database everyone uses.
Open supabase.com →Best practices for Claude Code
Plan mode, context hygiene, and step-sizing from the tool's makers.
Open code.claude.com →How Claude Code is used in practice
Data on what workflows actually succeed across thousands of users.
Open anthropic.com →Spec-driven development with AI: get started with a new open source toolkit
Why GitHub built the antidote to piecemeal prompting.
Open github.blog →Diving into spec-driven development with GitHub Spec Kit
A hands-on walkthrough of the spec-first loop.
Open developer.microsoft.com →EnzeD/vibe-coding: the vibe coding kickstart guide
A living, battle-tested prompt methodology with real example projects.
Open github.com →From prompt to prototype: get the most out of Lovable
A designer's method: describe design intent in steps, not specs.
Open medium.com →The non-technical PM's guide to building with Cursor
How a non-engineer at Meta structures work so Cursor delivers.
Open lennysnewsletter.com →Prompting hacks for vibe coding in Lovable
Small phrasing changes that visibly change output quality.
Open prettypromptai.substack.com →LLM development workflow example
Sees Harper's workflow applied to a concrete project end to end.
Open zacharynielsen.com →GitHub Spec Kit
An open-source scaffold that forces spec, plan, tasks, then code.
Open github.com →More in Vibe Coding
- Why can non-technical founders now build software with AI? 18 →
- What tools should I start with (Cursor, Claude Code, Lovable, Replit, Bolt, v0)? 19 →
- How do real founders ship an MVP with AI in days instead of months? 18 →
- What are the limits, when do I still need a real engineer? 20 →
- What actually counts as vibe coding, and how is it different from serious AI-assisted engineering? 18 →
- How do I keep my vibe-coded app from getting hacked? 19 →
- How do I get my vibe-coded prototype ready for real users? 18 →
- How do I add accounts, a database, and payments to my vibe-coded app? 18 →
- What do I do when the AI gets stuck, loops, or breaks things it already built? 18 →
- How much does vibe coding really cost, and how do I stop burning credits? 18 →
- Do I still need to learn to code, or is prompting enough? 17 →
- Can vibe-coded products become real businesses? 19 →
- Can I vibe code a mobile app and get it into the App Store? 18 →
- How do I avoid losing work? Version control for founders who don't code 19 →
Also in Starting Up
The same ground, over in Build the product, our Starting Up track.
eChai Partner Brands
