TL;DR: If you keep re-explaining a work version and a personal version of yourself to AI, the fix is not better prompts - it is two separate, portable contexts you own. Point each assistant at the right one, and both setup paragraphs disappear.
You have probably felt this without naming it. You open an AI chat for work and spend the first few minutes explaining your role, your team, and how you like things written. Later you open another chat for something personal, and you describe a completely different version of yourself from scratch. Same you. Two setups. Both typed again, every time.
The usual advice is to write better prompts. That is not the problem. The problem is that you are doing the setup work twice, because your assistant has no idea that your work self and your personal self are two different contexts that should not bleed into each other.
Why can't one assistant hold two versions of you?
Your AI keeps one running picture of you. Whatever you tell it for work and whatever you tell it for your personal life land in the same place. So one of two annoying things happens.
Either the two leak into each other - your clipped work tone shows up in a message to a friend, or a personal side project wanders into a work answer where it does not belong. Or you keep them apart by hand, re-explaining the right background at the start of every single chat so the assistant does not mix them up. That second option is the setup tax, paid twice.
Neither is your fault. You are managing a boundary the tool was never built to understand.
Does the low-tech version work?
There is a simple version that works for a while. You can keep two short documents, one for work and one for personal, and paste the right one in at the start of a chat. If you only ever use a single assistant for a single purpose, that is genuinely good enough, and you do not need anything more.
It breaks the moment your life gets slightly more complicated. Add a second assistant and you are keeping the same two documents in sync across tools. Let them drift and the AI is quietly working from a stale you. And on a busy day, "paste the right context first" is the step you skip, so you are back to re-explaining from memory anyway.
Why is built-in memory not enough?
The memory built into Claude or ChatGPT belongs to them. It lives inside their product, on their terms, and you cannot lift it out, split it, or point it at a different assistant.
Your context should be the opposite of that. It should be material you own: something you can keep in separate pieces, edit, switch off, and carry to any assistant you use. Owned and portable, not locked inside one company's app.
How do you keep work and personal apart, properly?
The fix is to stop storing your context inside a single assistant and start owning it as separate, portable pieces you control.
- Keep two contexts, not one blurred profile. A work context and a personal context, held as distinct things that you own.
- Decide what each one holds and who can see it. This is about control, not secrecy: you choose what goes in, what is shared, and what you can revoke later.
- Connect once, and point each assistant at the right one. The same context can be read across Claude, ChatGPT, Gemini, and Perplexity, so you set it up a single time instead of in every window.
That is the difference between an assistant that quietly mixes up your two lives and one that knows which version of you it is talking to.
Start small. Connect with Claude first, point it at your work context, and ask your real question without the setup paragraph. Then do the same for your personal side, kept separate and on your terms. This is exactly what UseMyContext.ai is for: context you own, kept in the pieces you decide, that any AI can read.
You were never bad at this. You were just doing the introduction twice. Own your context.
FAQ
Can I keep separate context for work and personal use?
Yes. The point is to hold your work context and your personal context as two distinct things you own, rather than one blurred profile inside a single app. You decide what goes in each and which one an assistant reads at any moment.
Does the memory built into my assistant not already do this?
Not really. The memory inside Claude or ChatGPT belongs to that product and lives on its terms, so you cannot split it cleanly, move it, or point it at another assistant. Owned, portable context is the part that lets you actually separate work from personal and carry both wherever you go.
Which AI assistants can read the same context?
Any assistant that supports MCP, the open standard for connecting context to AI. Today that includes Claude, ChatGPT, Gemini, and Perplexity. You connect once and they read the context you chose for them.
Can I control what each assistant sees?
Yes. You choose what each context holds, what is shared, and you can switch off access whenever you want. The goal is control you can see and change, not silent memory you cannot reach.