Juan Andrés Núñez.

Senior Frontend Engineer specializing in Agentic AI, professional educator and speaker. Where engineering meets humanism.

Juan Andrés Núñez
Focusing on

Building FrontendLeap: teaching how to build the professional Web with GenAI and learning science.

Helping companies and teams go AI First on the Claude platform.

Claude Code

✦ New tips daily

Claude Code: from zero to professional. Tips, tutorials and deep dive guides from an engineer who uses it 12 hours a day.

Explore the tips
Claude Code Tips View all →
#181

Claude Code token usage: you pay for the whole conversation again on every turn

I measured three real Claude Code sessions. For every new token I gave it, it re-read between 16 and 61 tokens of history. That, not the model, is what your session costs.

#180

Truncated output in Claude Code: the test failure that never makes it back

Your test suite prints 74,000 characters, then fails. Claude reads 10,000 of them, and the assert that broke is not in there.

#179

When to compact in Claude Code: before you step away, not when you get back

/compact doesn't cost the same at 12:55 as it does at 2:30. Same command, same conversation, and the price between them is ten times apart. What changed isn't the summary: it's whether your cache is still alive.

#178

Claude Code auto mode rules: write your own in one plain sentence

The classifier that decides what runs reads prose, not patterns. You can hand it a rule of your own, written the way you would explain it to someone joining the team today. And there is a nine-character token you cannot forget to include.

Gravitas

Get only what matters

If I have nothing worth saying, you won't hear from me. When I do, you'll be the first to know. 7,000+ people already trust this.

Are you a professional Web developer?
No

Unsubscribe at any time.