Comment by nadis

Thinking a lot lately about how AI is changing software development, and what the ideal development experience might actually look like in the future.

I’m working on CodeYam. We just recently released a lightweight CLI and a feature called CodeYam Memory. The immediate motivation was that when using Claude Code, the AI would repeat the same mistakes and our claude.md files would get stale too quickly to manually maintain. The existing options, including Anthropic’s built-in memory features, didn’t really solve the memory challenge sufficiently for us.

CodeYam Memory runs a background agent that reviews coding session transcripts, looks for patterns of confusion, and generates targeted rules with proper scoping.

It’s a small first step toward the longer-term idea I'm exploring: what an AI-native development experience should actually look like. For now, it’s packaged as a lightweight CLI you can use anywhere you’re running Claude Code.

How to try it:

Install: npm install -g @codeyam/codeyam-cli@latest

Then, from your project root run: codeyam

This will launch a dashboard with further instructions for initializing CodeYam Memory.

Free, runs locally, no login required, and language agnostic. Would love feedback.

More context:

Background blog post: https://open.substack.com/pub/codeyam/p/introducing-the-code...

90 sec demo on our own repo: https://youtu.be/oJ2gTb-lxbE

Demo teaching Claude a real OSS repo (Plane): https://youtu.be/CjOKBwBCcOs

Website: https://codeyam.com/