Adity Kushawaha

Founder of MemoryOS

About

Building MemoryOS and exploring how AI systems can keep reliable context as they learn, change, and work across multiple agents. Interested in AI infrastructure, agents, developer tools, and startups. Currently getting MemoryOS ready for launch and always happy to connect with other builders.

Links

MemoryOS

Twitter

Badges

Tastemaker

Gone streaking

Gone streaking 5

Forums

•

1mo ago

How do your AI agents handle outdated or conflicting context?

Hi makers

I m building MemoryOS, and I d love feedback from people working on AI agents.

Most AI products already store context. What interests me is how they govern it as things change:

A user corrects an earlier preference.
Two sources provide conflicting information.
Several agents update context about the same person.
Permissions change, but previously learned context still exists.