Your AI is smart. So why do you have to explain yourself every single time? đź§
Hi! This is the first post after the launch! I do see a lot of folks started creating their brains and putting it to work by connecting them to their agents!
Here’s the back story why i created this:
Every time I switch between Claude, ChatGPT, Cursor, or another AI tool, I run into the same problem.
The context doesn’t come with me.
My preferences, project details, decisions, and all the little things I’ve already explained? Scattered across different tools.
So I started building Make0 Brain - a persistent memory layer that lets your AI tools share context.
The goal is simple: teach your AI once, and stop starting from zero every time you switch tools.
We’re also publishing our benchmark results for LoCoMo and SWE-ContextBench because I wanted to evaluate the system beyond just saying it works.
But I’m curious about something:
If your AI could remember one thing about you permanently, what would you want it to remember? Your preferences? Your work context? Your ongoing projects? Something else entirely?
I’d love to hear how other people are dealing with AI memory today.
Replies
What I'd want remembered: my tone. I've told every tool I use that I write casually and they all default to corporate speak the second I start a new thread. That alone would sell me.
@paisley_coleman This problem is easily solvable via common memory store and Make0 Brain can work out really well for you!
This feels like the thing that should've existed two years ago. The fact that I have to reintroduce myself to a machine that just helped me debug for three hours is genuinely absurd when you say it out loud.
@owen_simmons That's a huge problem for everyone.
Okay but the real test is whether it remembers across model upgrades. If I set everything up today and GPT-6 drops next month, does my brain survive the jump or do I start over? That's the question nobody answers upfront.
@connor_coleman Yes, it is independent on model that you are using. Model can change but your memories will stay. Connect your brain to any AI tool like Claude, ChatGPT or even any future agent that may come.
I would want remembered permanently is the reasoning behind previous decisions rather than just the final answers. Is Make0 Brain designed to preserve that kind of context?
@jeankang Yes it does preserve that context and also has built in conflict resolution of memories and memory versioning.
Because most of the context lives in our heads, not anywhere the model can reach it. In the teams I've worked with, what changed things was treating documentation as the operating system for the agents: high-level design, low-level detail, guardrails, acceptance criteria, and what failure looks like. Writing it feels slow at first, but once it exists you stop re-explaining and start reviewing. Do you keep that context in the tool or in your own repo?
@bbueno You can dump those files, guardrails etc in the brain and your agent can access them via MCP. Your brain keeps on growing as agents read and write to your brain. It stops the need for creating md files ever again.