Rav

Rav

AI Agents Work. Their Memory Doesn't.
#102623201 follower0 following
✨ 3 day streak
>10,000All time
6 KP

About

20+ years in information security for global banking, securing systems that move trillions. 6 patents across Web2 and Web3. The newest failure surface I see is memory. Agents are deployed on systems that hold no verified state: no record of what is currently true, who committed to what, or what superseded what. Every trust framework I review assumes that record exists. Almost nowhere does it. So I'm building MrAgentˣ: the context layer for humans and their agents. Save your work in Claude, resume it in ChatGPT, never re-explain. What you received and what your AI concluded stay on separate tracks, never mixed. I write The Context Layer, a weekly series on the missing state layer, and post daily on AI agent security as @_MrDecentralize.

Badges

Gone streaking
Gone streaking

Forums

•

11h ago

From securing trillions in banking to fixing AI's memory problem

Hi Product Hunt. I'm Rav.

For 20 years my job was securing systems that move trillions of dollars in global banking. 6 patents, a lot of time spent on one question: why do trust models that pass design review break at scale?

•

1d ago

What are you building?

Comment below what you are building.

I'll start:

https://devdoc-ai-five.vercel.app/

It is an ai tool that will pull relevant content from pdfs or documentations the user gives e.g stripe docs given and stripe setup returned

•

2d ago

What are you building?

Doing one of these because I get more useful ideas from seeing what other people are building than from another generic marketing thread.

I'll start.

We're building Content Drifter, an AI social media workspace for founders, creators, marketers, and small teams.

It helps turn ideas, blogs, and videos into on-brand content, schedule it, publish across 9 platforms, and learn from real engagement so the next batch gets better.

View more