A little while ago I started using Claude more than other AI models. It was a pretty big moment for me considering how it happened Google Gemini was my previous go to LLM for all my needsbut I found that Claude did some tasks better than others. So I decided to use Claude as the AI to see where it excels and if it would be better to use a different AI.
One feature I quickly fell in love with was Claude’s memory tool. At first I turned it on so that Claude would get an idea of who I was and better tailor his answers to me. But as I used it, I felt that Claude was telling me things about myself that I didn’t even know.
Claude’s memory is very strong
I didn’t expect it to be this good
If you haven’t tried it before, Claude has an extra memory feature. If you activate it, Claude starts analyzing everything you discuss with him daily and records them in his memory. The more you use it, the more it records. The more it records, the more it knows about you, your needs, and the kind of person you are. So start every conversation with “Hey Claude, I’m XYZ;” there will be no need to start with you simply say who you are and it will carry that context into future conversations
The best part about this feature is that you can actually see what Claude remembers about you. You just open the settings page, open its cache file and see the little “user diary” it writes about you. If Claude gets something wrong, you can tell him to correct something in his memory and it will be fixed. Likewise, if you and Claude hit on a topic or idea you want him to remember, you can tell him to store it in his memory bank and he will.
When I first activated Claude’s memory tool, I expected the AI to store very simple, bulleted facts about me. You know, things like my name, projected age bracket, interests, etc.; There is nothing more complicated than a dating profile. However, the more I used the AI, the richer the memory file became, until it was strangely detailed and deep. The best way I can describe the feelings is that I read my personal assessment of myself, including who I am, what I like, and how I act.
Claude turns your idle questions and suggestions into self-reflection
He created a picture of me that I never knew existed
So here’s the thing. I’m not trying to pretend that Claude is some kind of super-intelligent computer that can analyze someone like a psychiatrist. When LLMs remember information about you and respond in an appropriate tone, they are reflecting on you and how you behave. They don’t take and dwell on what you put into it to create a picture of who you are; they only know what you feed them and use the information you give them to ask you more questions.
However, that doesn’t mean that Claude can’t teach you a little bit about yourself that you didn’t know. After all, its memory feature is aimed at remembering things that people don’t give you feedback on in real life. You have bosses to evaluate your work, and friends and family can teach you what kind of person you are, what you like and what you don’t like. However, LLMs can remember every little suggestion you give them, from your interesting questions to your full-scale projects, and use it to build a profile about you. Again, this profile is essentially just a mirror of you, but that doesn’t mean it can’t highlight things about yourself that you didn’t see before.
When I open my archive, I see that every suggestion and discussion I’ve made has been saved in a “Simon Report”. I don’t even remember some of the instructions I gave until I read the report, but Claude did. And since Claude remembered every suggestion I made, it could tell me something about myself that I didn’t know. For example, what Claude says about me:
His workflow is iterative and focused on precision, favoring technically precise and defensible language over sharp but imprecise statements. He reads the original sources and accepts corrections.
(…)
He has a pattern of intellectual interest in a variety of topics—linguistics, legal foundations, neuroscience, geopolitics, sleep science—and engages analytically, building understanding through follow-up questions and relating new information to personal experience.
The rest of the memory file covers my personal and professional life, offering me insights into how I deal with challenges, how I work best, and what I want to work with. All of this is information that I have deliberately fed, but Claude finds patterns in my behavior that I never recognized in myself. And that’s really cool.
Claude remembers the parts you forget
That’s because Claude pays attention to every little thing you do, from your serious business challenge to asking a random question, which allows him to produce such a glowing report of what kind of person you are. To be honest, reading the assessment that an AI made of you can teach you something that other humans can’t replicate.
