Mark Mwangi logoMark
Mwangi
Return to home
Get in
touch
Admin
login
ToolingJuly 25, 202610 min read

Tools every developer should know

The everyday toolkit — version control, terminal, package managers, API clients, and containers — that makes development reliable and fast.

By Mark Mwangi
Tools every developer should know cover coverImage

You don't need 50 tools to be effective, but a handful of well-understood ones remove friction from almost every task. Here are the categories of tools I consider essential, with concrete examples you can adopt today.

1. Version control: Git

Non-negotiable. Git tracks history, enables collaboration, and gives you a safety net.

bash
git status
git switch -c feature/login
git add -p
git commit -m "add login validation"

Pair it with a host: GitHub, GitLab, or Bitbucket.

2. The terminal

A comfortable shell (bash, zsh, or fish) beats clicking through GUIs for repetition and automation. Learn pipes, grep, sed, and how to write small scripts.

3. Package managers

  • **Node**: npm, pnpm, or yarn
  • **Python**: pip, uv, or poetry
  • **System**: Homebrew (macOS), apt (Linux)

Pin versions and use lockfiles so builds are reproducible.

4. API client

Tools like Postman or Insomnia (or plain curl) let you inspect requests, save collections, and debug endpoints without a UI.

bash
curl -X GET https://api.example.com/users/1 \
  -H "Authorization: Bearer $TOKEN"

5. Containers: Docker

Docker makes environments consistent ("it works on my machine" stops being a problem). A simple Dockerfile and docker-compose.yml can spin up your whole stack.

6. Editor + extensions

VS Code (or your editor of choice) with linting, formatting, and language tooling. Memorize its shortcuts — they compound daily.

7. Database client

TablePlus, DBeaver, or psql for inspecting and querying data directly instead of guessing from the API.

8. Observability

  • **Logs**: structured logging you can search.
  • **Metrics**: Prometheus/Grafana or your host's dashboard.
  • **Error tracking**: Sentry or similar.

9. AI assistants

Used well, an AI assistant accelerates boilerplate, explains unfamiliar code, and drafts tests. Keep it as a copilot, not an authority — review everything it produces.

A minimal starter kit

If you set up only a few today, pick: Git + a good terminal + your editor + Docker + an API client. That combination covers the majority of day-to-day development and scales from hobby projects to production teams.

DockerDevelopmentBest Practices
Go back to all blogs →Go to all projects →Contact →

Related Articles

  • How to solve merge conflicts
  • it works on my machine
  • HTTP methods

About

Mark Mwangi

Creating secure and meaningful digital experiences.

Software developer focused on secure, responsive, and high-quality digital experiences.

Socials

GitHubLinkedInXE-mailinstagramfacebookwhatsappMedium

© 2026 Mark Mwangi | All rights reserved

Terms of Service•Privacy Policy

Last updated: April 30, 2026 16:23:51 UTC