15:I[981,["658","static/chunks/658-a698693f56d0116a.js","183","static/chunks/183-6beeeb801c274aaa.js","474","static/chunks/app/%5Btype%5D/layout-631c4730b83c8223.js"],"Posts"] 16:T4e6,With O(n) memory, scaling 100,000x should require 100,000x more RAM.

I scaled from 1,000 to 100,000,000 tasks. Memory stayed at ~3GB the entire time.

Here's the cryptographic proof.

The benchmark: - 1K tasks: ~3GB memory - 100K tasks: ~3GB memory - 1M tasks: ~3GB memory - 10M tasks: ~3GB memory - 100M tasks: ~3GB memory

Same memory. 100,000x scale. Merkle-verified.

Hardware: Intel i7-4930K (2013), 32GB RAM

The proof: Every task is SHA-256 hashed into a Merkle tree. Root hash commits to all 100 million tasks. You can verify individual samples against the root. The math either works or it doesn't.

Root hash: e6caca3307365518d8ce5fb42dc6ec6118716c391df16bb14dc2c0fb3fc7968b

What it is: An O(1) memory architecture for AI systems. Structured compression that preserves signal, discards noise. Semantic retrieval for recall. Built-in audit trail.

What it's not: A reasoning engine. This is a memory layer. You still need an LLM on top. But the LLM can now remember 100M interactions without 100M× the cost.

Background: Solo developer. Norway. 2013 hardware.

Looking for: Feedback, skepticism, verification. Also open to acquisition conversations.

Repo: https://github.com/Lexi-Co/Lexi-Proofs17:T47f,i built Botkit because i couldn’t find a good way to see what other people are actually working on in robotics. a lot of projects seem to be scattered across Discords, X, private chats, or never shared at all.

Botkit is a simple experiment: you add a WhatsApp number and post project updates (text, photos, videos) while you’re building, like you’re messaging a friend. those updates become a project others can follow.

you can also forward purchase receipts and it’ll automatically extract the parts used, mostly so people can learn what actually goes into real builds. here's an exam

Happy Festivus – air your grievances

Dutch Firm Mollie Buys UK's GoCardless (YC S11) for €1.1B

Unix V4 tape recovered: first ever Unix written in C runs again

100,000x scale, same memory – cryptographic proof of O(1) AI memory

Deplatforming Backfired

Q2REPRO: Drop-in open source replacement for the Nightdive Quake II remaster

Show HN: Opinara – Global polling with real-time map visualization

Sparse-ternary-fma – 23x faster AVX-512 kernel for FHE&AI, from my ePrint paper

Seedance 1.5 Pro, better than Kling 2.6. New SOTA image to video model

Show HN: DeckBuilder Single-file HTML slide editor (750KB, no build step)

Carbon Battery Technology

Start of /top4 movement for indie/small web

Investigating NVMe LBA sizes and formatting for performance (2023)

Issue with the New Asus Rog Matrix 5090 GPU

Dirlock

China Delays Plans for Mass Production of Self-Driving Cars After Accident

Show HN: Blobbing – a doodling experiment where dots become blobs

Show HN: Experimental) livestream anything, tree+mesh network

Training AI to do my investment analyst job

SMART_tests_with_smartctl

South Korea will require face scans to buy a SIM

Show HN: Botkit – Share robotics projects by posting updates via WhatsApp

Repath-studio: A local web-based vector graphics editor

Nanostructured coatings puncture bacteria to prevent biofilm formation

Another Way to Rate Incidents

Moravec's Paradox and the Robot Olympics

Show HN: Visual Countdown Timer (Open Source)

LoPA: Scaling Diffusion LLM Single-Sample Throughput to 1000 TPS

Business SLOs

Malicious extensions in Chrome Web store steal user credentials

Show HN: Automated PostgreSQL backups that verify they work