Author: Jose S
Why passing tests don’t prove ChatGPT Codex got the code right
A passing test only proves the cases it tries. When Codex writes the fix and the test together, check three things: the new test fails without the fix,…
Should you use Claude chat, Claude Cowork, or Claude Code for your job?
Chat, Claude Code, and Claude Cowork share a model family but not a job. Part 1 of the Claude Cowork tutorial separates conversation from software work from office…
How to automate data validation checks so bad data never reaches your reports
Automated checks catch bad data before leadership does. Build small scripts for row counts, referential integrity, freshness, and yesterday versus last week sanity so every number you ship…
How to use skills and scheduled tasks in ChatGPT Codex safely, one at a time
Skills, tasks, and scheduled help in ChatGPT Codex are power-ups, not day-one defaults. Climb a complexity ladder, hedge product renames, and never leave high-stakes deploys unattended.
Team habits that keep AI coding tools like Claude Code safe
Part 10 closes the Claude Code tutorial with team rails: shared CLAUDE.md and skills, required PRs and green CI, secrets policy, no silent prod deploys from agents, and…
