Aetna

Have a Safe Memory 🧠
AI Memory Specialist | audit | trust, focused on turning research into products.

[Aetna @AtMemX]

Memory is underrated part of your AI Agent.

[Aetna @AtMemX]

@devfrom_hyd H2

[Aetna @AtMemX]

Datacenter produces heat so why not boiling water and generating power sing micro turbines instead of fans at least noise pollution will benefit. Water then can leave in a closed loop 🔁

[Simran Tech @devfrom_hyd]

Why not build data centers in places where cooling is naturally easier and cheaper?

[Aetna @AtMemX]

@devfrom_hyd Datacenter produces heat so why not boiling water and generating power sing micro turbines instead of fans at least noise pollution will benefit. Water then can leave in a closed loop 🔁

[Aetna @AtMemX]

@SFourdrinier Indeed. But the complexity is the scope you try to optimise. Wider local optimal points compete. There might not be a global optimisation feasible… the aim shall be how to slice the task to give better chance to harness to engage correctly and get alignment right

[Aetna @AtMemX]

https://t.co/eaP8cktNNo

[Aetna @AtMemX]

https://t.co/mx4XrRrNTO

[Aetna @AtMemX]

“Subagents are not for playing house.”

Don’t build a fake company of AI researchers, managers and reviewers just to pass the same task around.
Use subagents when work can be divided into clear, independent tasks or run in parallel.
Subagents --> Control your context --> manage context.

[Aetna @AtMemX]

Caring about small biz is huge

[OpenAI @OpenAI]

Introducing ChatGPT Business Premium Seats

The new $100 Premium seat is a game changer for small businesses and startups—giving lean teams better tools, faster workflows, and capabilities once reserved for big companies.

A flexible plan that scales with your team’s ambition.

https://t.co/zxvwwaOaVF

[Aetna @AtMemX]

Tempted

[Tim Cook @tim_cook]

The new Mac mini is here. Small in size. Big on performance. From everyday productivity to all things AI, it can help you do it all.

[Aetna @AtMemX]

It’s good news well done 👍

[Sam Altman @sama]

we made a chip and it is fast

[Aetna @AtMemX]

@altrhorizon We have yearly subscription and we don’t get the value we thought we can get art signed up

[Aetna @AtMemX]

Do you agree sentiment around Anthropic is becoming more negative everyday?
I don’t buy the argument they don’t care about users they are after enterprise… I think these are related they will get a hit there too

[Aetna @AtMemX]

@gurtej__gill_ Thanks for feedback we need to hide some details and make it easier to follow

[Aetna @AtMemX]

Ok using this https://t.co/4oXOl87RYd for observation the first time was a smooth ride ..: still not comfortable with all the insights have you used it before?

[Aetna @AtMemX]

@vedanthk_ Better quality in implementation of complex logics workflow and no drama like Fable limit etc … indeed Claude still is better in design and front end but I will still prefer code quality and token

[Aetna @AtMemX]

@vedanthk_ Agree reset helps but it’s more I cannot believe in the last two weeks I have used personally only 10% of the time with Claude
Kinda don’t miss it anymore GPT replaced it

[Aetna @AtMemX]

https://t.co/GT6CG1vsns

[Aetna @AtMemX]

@an_engineer_log I think they spending more time “watermarking” the text than actually writing ✍️ it

[Aetna @AtMemX]

A lesson for early adopters of AI agents:

0. Build before you buy.
0. Run agents in real workflows, not just demos.
0. Treat failures as engineering problems: state, retries, memory, permissions, observability and recovery.
0. Test open-source models seriously.
0. Avoid locking into an infra stack too early.
Eat your own dog food.

This space is moving too fast to learn from the sidelines.