saba-ch

Anthropic recently launched Claude Tag, a new product that adds Claude as an AI Coworker in Slack.

It works by reading your conversations in channels while building a persistent memory of them. If the idea of always-on agent learning from internal conversation gives you a pause, or you just want to have it on your own terms, this post is for you.

We will be building a self-hostable version of it from scratch.

Yeah, not only complex but very expensive at scale, you would want to preindex or heavily delegate to smaller models.

There are a lot of code review tools popping up lately.

People assume these tools are very complex, while some might be, building one you can use on day to day basis is pretty simple and straightforward.

In this post I will take you through the whole process

Thank you for your feedback, that's what i am usually getting from people, it looks cool but they don't know what to do with it.

I am planning to add automatic analysis features for: cost, security, unused resources. I think it will add value to it and give users kick start.