Skip to content

AI Tools

AI tools have become a regular part of how we work at Planet Argon. From coding assistants and chat interfaces to IDE integrations and documentation helpers, our team uses a growing collection of tools across all disciplines. The landscape moves fast, and we are actively learning together as it evolves.

AI shows up at every stage of the work we do:

  • Planning — Exploring requirements, sketching out architecture, evaluating approaches, and thinking through database or API design.
  • Development — Writing and refactoring code, catching errors, generating boilerplate, and working through unfamiliar parts of a codebase.
  • Testing — Generating test cases, identifying gaps in coverage, and improving existing tests.
  • Code Review — Getting a second set of eyes on a diff, spotting potential issues, and drafting review feedback.
  • Documentation — Writing technical docs, summarizing decisions, and drafting communication for clients and teammates.
  • Debugging — Investigating errors, reading stack traces, and researching unfamiliar behavior.

These tools are meant to augment our work, not replace our judgment. We are still responsible for every line of code we ship, every PR we approve, and every recommendation we make to a client. AI helps us move faster and think more broadly, but the final call is always ours.

The AI tool landscape changes quickly. When evaluating whether to adopt a new tool, we ask ourselves:

  • Does it genuinely make me more productive, or is it just novel?
  • Does it fit into my existing workflow, or does it create more friction than it removes?
  • If it costs money, is the value worth it for how often I will use it?
  • Is there good documentation and an active community behind it?

Not every tool is the right fit for every person or every project. Use what works for you, and share what you find with the team.

All AI tools must comply with our internal security best practices as well as any privacy and security standards set by our clients. Be thoughtful about what you share with AI tools — do not input sensitive client data, credentials, or proprietary information unless the tool is specifically approved to handle it. When in doubt, ask.

This is where we get better together. AI tools are new enough that everyone is still figuring out what works. We encourage you to:

  • Talk shop in #engineering-lab — This is our Slack channel for sharing tips, interesting prompts, tool recommendations, and workflow ideas. Drop in what you are learning, even if it is rough.
  • Pair up — If you have found a workflow that is working well, offer to pair with a teammate and walk them through it. If you are curious about how someone else is using a tool, ask them.
  • Share your mistakes — Not every experiment works out, and that is valuable information too. If you tried something and it went sideways, share what happened so the rest of the team can learn from it. We would rather hear about a failed experiment than have five people make the same mistake independently.

AI tools are evolving quickly and so is our understanding of how to use them well. The more we share, the faster we all improve.

Claude Code has been approved for use on most client projects. We encourage every developer on the team to use it as part of their daily workflow.

We maintain a private repository of shared skills and rules for Claude Code at planetargon/pa-dev-skills. These skills help standardize how we use Claude Code across projects — from PR reviews to commit conventions. We encourage you to use the skills that are available and to contribute new ones as you discover useful patterns.

If you are new to Claude Code or want to sharpen your skills, check out Claude Code for Curious Rails Developers for a practical introduction.