maaz

Two execution traces can look completely different even when nothing important changed. Timestamps, random IDs, and event ordering create noise that makes real changes hard to spot. Built TraceDiff: feed it two traces, get back only what actually changed. Demo Link Below

I left the batter to ferment overnight, but now it smells bad.🥀

I interacted to one person and now my whole feed is just that person .

I am just reading about agentic memory and its not ending 😭

Do What Thou Wilt

I will dedicate these racks to my friend @infinterenders

I was going to build something like graphify in my 7th sem as the minor project but my mentor rejected it , they didn't like the idea lol

After adding CEO and co founder to their bios, these folks are posting anything these days .

????????

Just sleep bro

The algorithm loves visual typa shit because people can't understand words

"grok confirm this"

I know this lame ahh algorithm is not pushing my posts to my omfs , fk this algo

I don't usually write technical articles, and I'm pretty weak at this stuff. I don't do it often, and because of that, whenever I write one, I keep thinking about how I should structure it correctly, and I have to invest a lot of time. Well well, anything I do, I have to spend a lot of time on, ahh.

So the writing stuff I’m trying out, tbh, is for better visibility (I know 99% of people are not going to read it still), and I will try to post only the stuff I actually know, to make full use of the blue tick. The end goal, after reading about these products and research papers, is to build an agentic memory, and I didn’t know a lot about agentic memory, so I’m spending time doing a thorough review of what’s available, how they address the problems related to agentic memory, and what problems still remain to be solved. After absorbing this content, I will talk about how I will build my own agentic memory. Happy if you all follow along

Why is my reach so dead yarr 🥀🥀

Sometimes what happens is, I'm writing something, a word comes to mind, I write it down, and then I go look up its actual meaning 😂

why this disparity 🥀

Lately I've been digging into agentic memory , and came across an interesting way it is being classified based on functionality: Episodic memory - stores experiences and events. Semantic memory - stores facts, concepts, and general knowledge. Procedural memory - stores how to perform a task. Reflective memory - derives lessons from past experiences and mistakes. These aren't necessarily four distinct types of memory in every architecture. They're different aspects of how an agent stores, processes, and uses information, and they can work together to form a memory system. What's interesting is that each aspect comes with its own set of challenges that researchers are actively trying to solve.