Ivalasa🏴‍☠️

🌍 Reflecting on tech, society, and the future | Satire & humor | Social Media Pro | 🏴‍☠️ Digital Pirate

[Ivalasa🏴‍☠️ @ivalasa]

@santander_es que tardéis en solucionar el fallo ?

[Ivalasa🏴‍☠️ @ivalasa]

@MonchoMan @santander_es Si , sigue fallando

[Ivalasa🏴‍☠️ @ivalasa]

@AlertaNews24 Adidas Rompe tobillos por 500$

[Ivalasa🏴‍☠️ @ivalasa]

@fcoury No reset..

[Ivalasa🏴‍☠️ @ivalasa]

@thsottiaux Not for me 😭

[Ivalasa🏴‍☠️ @ivalasa]

@thsottiaux It's not working for me 😔

[Ivalasa🏴‍☠️ @ivalasa]

Ctrl+C / Ctrl+V
That's coding today

[Ivalasa🏴‍☠️ @ivalasa]

@eliana_jordan Welcome to Spain

[Ivalasa🏴‍☠️ @ivalasa]

@thebeekeeperfox E scump,da...

[Ivalasa🏴‍☠️ @ivalasa]

@DavidOndrej1 How can I make these two work together? 🤔 They don't know UI

[Ivalasa🏴‍☠️ @ivalasa]

@bcherny How many millions of tokens?

[Ivalasa🏴‍☠️ @ivalasa]

@AishwaryaDevv The Pro plan takes me 5 prompts and out. ....

[Ivalasa🏴‍☠️ @ivalasa]

@andreibratucu @MarioNawfal @nikitabier Crypto , not rubles 4

[Ivalasa🏴‍☠️ @ivalasa]

@sama Codex

[Ivalasa🏴‍☠️ @ivalasa]

Codex from @ChatGPTapp better than @claudeai...

[Ivalasa🏴‍☠️ @ivalasa]

@BenidormAyto Pinos? !?

[Ivalasa🏴‍☠️ @ivalasa]

@melvynx Fals

[Ivalasa🏴‍☠️ @ivalasa]

@thugGauss Yo me he cambiado a chatgpt

[Timmy Ghiurau @itzik009]

After 10 years at Volvo Cars, I left to co-found MidBrain.

AI doesn’t fail because it lacks intelligence.
It fails because it doesn’t remember.

Over the past year, we’ve explored this across very different environments:

-AI agents in Minecraft
-Long-running AI companions
-Human behavior simulation in robotics
Different domains. Same failure mode.

AI agents process information -
but they don’t improve from experience.

Today, most AI systems are:
Stateless across sessions

Dependent on large context windows

Repeating the same mistakes
Retrained in expensive batch cycles

They retrieve the past.

But they don’t learn from it.

What’s often called “memory” today is:
store → retrieve → inject → re-run
This improves context.
It does not change behavior.

At MidBrain, we’re building the missing layer:

Memory + continual learning for long-running AI systems

Our thesis:
Intelligence is not just inference.
It is the ability to change through experience.

Today, we’re sharing our first step:

SmartSearch - memory retrieval for long-horizon agents

93.5% on LoCoMo (SOTA)

88.4% on LongMemEval-S (SOTA)

LLM-free retrieval (CPU-only)
~8.5× fewer tokens

Paper: https://t.co/63THNXAFcS

This work shows:

why retrieval alone doesn’t lead to learning
why memory must consolidate into procedural behavior

why today’s training paradigm is fundamentally inefficient
(2–5× wasted compute, repeated data movement, batch retraining loops)

This is the starting point.

From:
retrieval → memory → learning → behavior

Toward:
AI agents that run for years, not minutes
One identity across chat, code, and physical environments

We’re opening early design partnerships
for teams building long-running agents (copilots, coding agents, companions)

https://t.co/qzj2gX7WPh

[Ivalasa🏴‍☠️ @ivalasa]

@gbuhnici