Show HN: Memoram – A user-owned memory layer to personalize any AI tool | Hacker News Reader

Show HN: Memoram – A user-owned memory layer to personalize any AI tool

memoram.app

3 pointsCodyGaskin4 comments

Hey HN,

We've developed Memoram, a platform that allows users to securely store and manage their personal data—like preferences, conversations, and other relevant information—across various AI tools. This ensures a consistent and personalized experience, regardless of the AI application in use.

Key Features:

Secure Data Storage: Users can store their data with end-to-end encryption, ensuring privacy and security.

Cross-Platform Integration: Memoram integrates with multiple AI tools, allowing seamless access to user data across platforms.

User-Controlled Sharing: Through unique MemoryKeys, users have granular control over which AI tools can access specific data.

Developer Platform: We've launched a Developer Platform that provides APIs for developers to integrate Memoram into their AI applications, enhancing personalization without building user data management from scratch.

We're eager to hear feedback from the community, especially from developers working on AI applications. How do you currently handle user personalization? Would a centralized, user-controlled memory system like Memoram be beneficial in your projects?

Looking forward to your thoughts and discussions.

Open articleView on HN

This sounds disingenuous and misleading. There is no way LLMs can be cajoled into consistent results. It's all prompts, all the way down.

While this service may add more prompts around prompts around prompts, that's all it can do. Cursor and other tools already do that with rules files.

It does not help the overall case for AI tools when we promise things they can't deliver or mislead people about how they work. Using words like "learning" and "memory" here in your advertising are examples of that.

Thank you for your comment and for raising concerns about the accuracy of our messaging. I’d like to clarify what Memoram actually provides, as it’s fundamentally different from just “prompt engineering” or stacking prompts.

Memoram is not an LLM or an AI tool itself. Instead, it’s a secure, encrypted database where you can store any information you choose—preferences, facts, notes, or other data you want to persist across AI tools. Think of it as a personal, cross-app “memory vault” that you fully control. When you interact with different AI assistants, you can selectively grant them access to relevant pieces of your stored information, enabling those tools to provide more context-aware and consistent responses, regardless of which LLM or service you’re using

Thanks for the reply and I can see how that might be helpful. It would definitely improve the pitch to say that directly and leave out the statements that it improves understanding or results.

Thanks. I will certainly make it more clear.

Reply on news.ycombinator.com