How I use Claude to build software3 min
AI · Claude
Working with Claude
Claude is the AI I spend the most hours with — for engineering, writing, and running agent systems. This hub collects what I've learned about getting consistently good work out of it.
The short version: treat Claude like a talented collaborator who's new to your context. The quality of what you get tracks the quality of what you give — context, constraints, and honest feedback.
The essentials
Give it your context
Paste the real document, the real code, the real constraint. Vague questions get plausible answers; specific inputs get useful ones.
Iterate like a colleague
Don't restart when output misses — say what's wrong and why. Claude improves sharply on honest, specific feedback.
Graduate to projects and tools
Standing instructions, project knowledge, and tool access (like MCP) turn Claude from a chat window into infrastructure.
Latest on this
All writing →The Build Log
The stuff I learn while building.
I test a lot of tools, workflows, and ideas. The Build Log is where I share what actually worked, what surprised me, and what I would do differently.
AI workflows. Claude. Agents. Building software. Building companies.
- Practical AI workflows I actually use
- Tools and systems worth your attention
- Lessons from building software and companies
- Things I learned the hard way
One email when there is something worth sharing. From Charles, no one else.
