Technical
The Tools That Changed How I Build
A year ago my toolbox was bloated. Every week I tried something new. Every month I abandoned half of it. The business problem is not a shortage of AI tools. It is the opposite. Choosing the wrong ones costs time, trust, and money. Consultants and founders who pick well ship. The rest get stuck in demos.
After 365 days of writing and building, only a few tools earned a permanent seat. I want to be honest about which ones changed how I work and why.
The short list
- Claude Code for anything multi-file. It respects context. It refactors across a repo without losing the plot.
- Cursor for a tight inner loop. Tab completion that actually understands my codebase.
- Plain Python scripts run by an LLM. No framework. Just
urllibandjson. - Vercel because deploying a Next.js app should take thirty seconds, not thirty minutes.
- DynamoDB single-table design for most serverless apps. One table, many patterns.
Why these stuck
They share one trait. Each one reduces a decision I used to agonize over. Claude Code decides where a function belongs. Cursor decides what the next three lines are. Vercel decides my deploy pipeline. The win is not raw speed. The win is fewer forks in the road.
A small example
# The whole deploy script I run most days
import subprocess
subprocess.run(['vercel', '--prod', '--yes'], check=True)That is the script. Three lines. No CI config. No Docker. That simplicity is the point.
What I dropped
Anything that added a layer without removing a decision. I dropped two agent frameworks, one vector database, and a workflow tool that promised orchestration. They added ceremony. They did not add clarity.
The rule
If a tool does not collapse a decision tree, it is noise. See the Claude Code docs and the Vercel CLI reference for the two I rely on most. Keep the soup small. Keep the spoons sharp.
RELATED READING
The Consulting Shift I Am Making In Year Two
After a year of writing and building, my consulting practice is changing shape. Shorter engagements. Sharper outcomes.
ReadThe Frontend Shift: Shipping Less JavaScript In Year Two
A year ago I reached for Next.js for everything. This year I often reach for nothing.
ReadThe Serverless Lesson I Would Write On A Sticky Note
After a year of shipping serverless projects, one rule explains most of the wins and all of the losses.
Read