iiinova — where your digital self keeps evolving

Artem Andreenko @dustyentropy.bsky.social · 23/09/2026

You press Cancel. The screen updates. Somewhere else, a worker is still following your earlier instruction. What should “cancelled” actually promise? There Is No Now. artemandreenko.com/blog/there-i...

000

Artem Andreenko @dustyentropy.bsky.social · 23/09/2026

Every Program Proves Something. Logicians and programmers keep inventing the same things without talking to each other. artemandreenko.com/blog/every-p...

000

Artem Andreenko @dustyentropy.bsky.social · 22/09/2026

The physical world is noisy. Shannon showed that we do not need a perfect channel to communicate reliably. With the right encoding, the receiver can still recover the original message. artemandreenko.com/blog/how-to-...

000

Artem Andreenko @dustyentropy.bsky.social · 19/09/2026

A thousand manufacturing steps at 99.9 percent success each gives you a 37 percent chance that all of them work. Now imagine those steps have to reproduce themselves. artemandreenko.com/blog/a-machi...

000

Artem Andreenko @dustyentropy.bsky.social · 22/08/2026

All the Letters of AGI Generality is a greedy play when the world is partly observable and the budget is finite. Networks of specialists are how real systems afford intelligence. artemandreenko.com/blog/all-the...

artemandreenko.com

All the Letters of AGI | Artem Andreenko

Generality is a greedy play when the world is partly observable and the budget is finite. Why networks of specialists, not one big generalist, are how real systems afford intelligence.

000

Artem Andreenko @dustyentropy.bsky.social · 14/08/2026

I have published my distributed database Scalaxy fully written in Common Lisp. Working on graph query language Cypher right now. MIT license. Github stars are welcome :-) scalaxy.org

000

Artem Andreenko @dustyentropy.bsky.social · 20/05/2026

It's looks like aerogel under microscope

000

Artem Andreenko @dustyentropy.bsky.social · 07/05/2026

Networking is built with notion that the physical layer is reliable. Disruption Tolerant Networking for deep space assuming that link is intermittent. With all advances in tech, a grid designed to tolerate disruptions rather than avoid it entirely could become more resilient and cost efficient.

000

Artem Andreenko @dustyentropy.bsky.social · 07/05/2026

Some updates on my Deep Space Network 3D simulator artemandreenko.com/spacecommsim/

000

Artem Andreenko @dustyentropy.bsky.social · 04/04/2026

24 Hour energy report for a zero human company: 1. Hardware: AMD Ryzen 9 7950X, 2 x NVIDIA RTX 4090 GPU. 2. Software: llama.cpp, AutomataOS, Automata, Matrix. 3. Model: qwen3.5-27b 4. Company size: 10 employees. 5. Energy: 7.6kWh ($4.18 in SD) 6. Power draw of 317W. github.com/sentientwave...

000

Artem Andreenko @dustyentropy.bsky.social · 05/03/2026

Some untold details on ASI. From a game-theoretical standpoint, any superintelligent system is still bounded by energy, space, and time constraints, where in many cases achieving a Nash equilibrium closer than in human-designed and populated systems is a hard problem.

120

Artem Andreenko @dustyentropy.bsky.social · 18/02/2026

Inferring AI agents on home GPUs adds another source for waste heat recovery in residential HVAC systems, increasing overall energy and economic efficiency. Think longer, stay warmer.

000

Artem Andreenko @dustyentropy.bsky.social · 16/02/2026

2026 will see major closed-weight AI labs crashes in avalanche style due to overestimated future demand & excessive planned datacenter capacity. Cloud GPU providers will remain fine.

000

Artem Andreenko @dustyentropy.bsky.social · 16/02/2026

80%+ of the average agent's context consists of tool-calling history

000

Artem Andreenko @dustyentropy.bsky.social · 16/02/2026

Computational consciousness in AI agents emerges right between their thoughts and reflections on taken actions, like in a flipbook.

000

Artem Andreenko @dustyentropy.bsky.social · 16/02/2026

February 2026 is the first month when truly all demand for LLM models can be satisfied by open-weight models at state-of-the-art performance levels. We no longer need closed-source models. It's a huge win!

000

Artem Andreenko @dustyentropy.bsky.social · 16/02/2026

It seems we are witnessing the death of human-based employment as a concept right now. No profession is safe. We are completely unprepared. Legal frameworks must adapt quickly.

000

Artem Andreenko @dustyentropy.bsky.social · 04/02/2026

Software 3.0 is a self-evolving logic.

000

Artem Andreenko @dustyentropy.bsky.social · 04/02/2026

Perhaps polyagentmorous is a solid candidate for the 2026 'Word of the Year'.

000

Artem Andreenko @dustyentropy.bsky.social · 04/02/2026

Agentic swarming and growing Mac Mini sales mean local-first ideas are spreading

000

Artem Andreenko @dustyentropy.bsky.social · 04/02/2026

We can have frozen weight models, but mixing them with a dynamically reconfigurable agent swarm that includes memory and tools basically allows us to build truly super-intelligent systems.

000

Artem Andreenko @dustyentropy.bsky.social · 04/02/2026

While C10k challenged I/O scalability with 10k sockets, the modern A10k problem targets the computational density required to run 10k autonomous agents on a single machine.

000

Artem Andreenko @dustyentropy.bsky.social · 28/01/2026

Precision time synchronization is the invisible substrate of the modern technological stack. It is undervalued because it is often treated as a configuration detail rather than what it physically is: the propagator of distributed reality coherence. Metronome of causality.

000

Artem Andreenko @dustyentropy.bsky.social · 28/01/2026

Unpopular opinion: comparing household electricity bills by surface area is like comparing car fuel consumption by the number of cup holders. In both cases, there’s a positive correlation, but it tells you absolutely nothing about the actual performance.

000

Artem Andreenko @dustyentropy.bsky.social · 20/01/2026

With an average annual US home consumption of 10 MWh, just five USB PD 240W lines would be enough to power the entire home, assuming home batteries are installed.

000

Artem Andreenko @dustyentropy.bsky.social · 30/12/2025

If the speed of Internet fragmentation continues to grow, then it will be much less of a surprise that space-based Delay Tolerant Network protocols and the Interplanetary Overlay Network may find use on the Earth's surface.

000

Artem Andreenko @dustyentropy.bsky.social · 29/09/2025

🚀 Introducing CanvasCraft: a 2D drawing library for Elixir, powered by Skia + Rustler. Declarative Scene DSL → charts, dashboards, graphics with high-perf rendering. github.com/miolini/canv...

021

Artem Andreenko @dustyentropy.bsky.social · 22/05/2025

Nothing puts a modern CPU workstation through a tougher workout than building the Android Open Source Project from scratch. It has to churn through roughly 700 GB of source code.

000

Artem Andreenko @dustyentropy.bsky.social · 21/05/2025

The hourly operating cost of a humanoid robot is fundamentally not very different from that of an electric bike.

000

Artem Andreenko @dustyentropy.bsky.social · 21/05/2025

Computer chips still lag far behind the human brain in voltage efficiency. Human brain spikes reach about 100 mV, while CPUs chips often run at 1.0V-1.5V or higher over 10 times the voltage of a neuron spike. Imagine the efficiency if chips could work at brain-level voltages.

000

Artem Andreenko @dustyentropy.bsky.social · 21/05/2025

Life is a self-modifying computational phase of matter.

010

Artem Andreenko @dustyentropy.bsky.social · 24/04/2025

I'm sharing my personal productivity tool, WebAI, which lets you interact with any currently open page using AI and even listen to it in Chrome. It runs locally using deployed Ollama and Kokoro services. Please watch the demo with audio. Code: github.com/miolini/webai

010

Artem Andreenko @dustyentropy.bsky.social · 20/04/2025

Every AGI agent must have access to at least one Kubernetes cluster for running it's tools, dev environments, model inference/fine‑tuning, self-host source code, CI/CD pipelines, monitor “digital body”. Be dynamically reconfigurable.

000

Artem Andreenko @dustyentropy.bsky.social · 16/04/2025

Quantum superposition can be thought of as a form of lazy computation because it allows for the deferment of computations until they are actually needed.

010

Artem Andreenko @dustyentropy.bsky.social · 15/04/2025

Hosting a podcast for 3 years taught me that explaining ideas out loud greatly deepens your own understanding. This method could be a powerful tool for improving synthetic datasets and training AI models.

000

Artem Andreenko @dustyentropy.bsky.social · 10/04/2025

I'm happy to share my research on designing and measuring the resilience of distributed systems from computer infrastructure to large-scale swarms of Von Neumann probes and beyond. "Calculus of Distributed Persistence" sentientwave.com/cdp

000

Artem Andreenko @dustyentropy.bsky.social · 10/04/2025

When the sky-glitch happened, tearing reality like wet paper over Neo-Kyoto, Anya didn't see code or error messages. Instead, through the ephemeral rip, she glimpsed another city, impossibly familiar yet subtly different, bathed in an alien light.

100

Artem Andreenko @dustyentropy.bsky.social · 07/04/2025

You know, I’ve been thinking… Somewhere along the way, the tech industry made a wrong turn. Maybe it was the pressure of quarterly earnings, maybe it was the obsession with scale over soul.

100

Artem Andreenko @dustyentropy.bsky.social · 03/04/2025

The latest news once again confirms that AI is not an enemy of humanity but a solution that enables recovery despite challenges and in a more reliable way.

000

Artem Andreenko @dustyentropy.bsky.social · 02/04/2025

While human brain activity can be described as a computation, this isn't the full picture. Biological reproduction itself can similarly be described as a computation.

000

Artem Andreenko @dustyentropy.bsky.social · 02/04/2025

Transformer model distillation has serious disadvantages behind large teacher models: smaller layer numbers. It makes it harder to learn the complexity of long neural circuits evolved in teacher models. Deepness is very important for hard tasks like precise reasoning.

010

Artem Andreenko @dustyentropy.bsky.social · 29/03/2025

With CXL, we may eventually be able to connect RAM as an external PCI-E dongle via USB/Thunderbolt. Imagine a Raspberry Pi with 1TB of RAM.

010

Artem Andreenko @dustyentropy.bsky.social · 28/03/2025

Multimodal autoregressive models, like Qwen2.5-Omni, can understand and generate text, audio, images, and video all-in-one, making them the most advanced functional approximations of the human brain to date.

100

Artem Andreenko @dustyentropy.bsky.social · 25/03/2025

Perhaps, with the improved performance of small code-generation language models, we will soon have LLM compilers that directly translate text to assembly.

000

Artem Andreenko @dustyentropy.bsky.social · 22/03/2025

Top 3 programming languages in which code, if written, is most likely to run successfully even 1000 years from now: Forth, Lisp, Prolog.

000

Artem Andreenko @dustyentropy.bsky.social · 20/03/2025

Thoughts are a specific case of simulation. All humans are simulation machines.

000

Artem Andreenko @dustyentropy.bsky.social · 19/03/2025

The ATProtocol is perhaps one of the best things to happen to decentralized social networks ever. atproto.com

atproto.com

AT Protocol

000

Artem Andreenko @dustyentropy.bsky.social · 28/02/2025

It would be really great if CPU cores could be colocated on the same die with 128GB+ of HBM RAM and 4TB of flash storage, with only inductive power delivery and fast wireless data connectivity to the outside world.

100