Second Brain 3.4.0: give your AI the right context for each project
We just shipped Second Brain 3.4.0, and since the last Product Hunt update was 3.0, I wanted to share what has changed.
The biggest addition in 3.4 is Projects.
If you use AI across different products, clients, codebases, or areas of your life, one giant pool of memory can get messy fast. Projects let you keep related memories together, so your AI can pull from the right context without you having to explain everything again.
You can create a project for something like:
• a product you are building
• a client you work with
• a research topic
• a codebase
• a long-running personal goal
Existing tags can also become project aliases, so memories you already have can be organized into projects without manually moving or rewriting them.
Since 3.0, we have also added:
• Prompt Capsules, so you can give AI tools a stable set of important context without relying on a fresh recall every time
• A redesigned dashboard, so it is easier to see what your Second Brain knows, what you come back to most, and how your memories connect
• Better team and integration controls, so you have more control over what stays personal, what is shared, and where connected sources save their memories
• Recall and graph improvements, so finding the right memory is more reliable as your Second Brain grows
The goal is simple: your AI should not just remember more. It should remember the right things for what you are working on right now.
With Projects in 3.4.0, Second Brain feels much closer to that.
Curious how others here organize long-term AI context today. Do you think about it by project, client, topic, or something else?


Replies
Dial
by client, mostly, but we learned that the hard way. early on we kept one shared memory pool across every customer's voice agent config and it seemed fine until two clients had similar sounding requirements and the agent started blending assumptions from one into the other during a call. nothing catastrophic, just a wrong detail stated confidently, which is almost worse than an obvious error. now each customer is its own boundary and topic/codebase context lives one level under that. curious whether your "existing tags become project aliases" migration is lossy at all, or does everything that used to be cross-tagged just end up belonging to multiple projects at once after the conversion?