The problem with LLMs isn't intelligence; it's amnesia and dishonesty. Hey HN, I’ve spent the last few months building Remember-Me, an open-source "Sovereign Brain" stack designed to run entirely offline on consumer hardware. The core thesis is simple: Don't rent your cognition. Most RAG (Retrieval Augmented Generation) implementations are just "grep for embeddings." They are messy, imprecise, and prone to hallucination. I wanted to solve the "Context integrity" problem at the architectural layer.

The Tech Stack (How it works):

QDMA (Quantum Dream Memory Architecture): instead of a flat vector DB, it uses a hierarchical projection engine. It separates "Hot" (Recall) from "Cold" (Storage) memory, allowing for effectively infinite context window management via compression.

CSNP (Context Switching Neural Protocol) - The Hallucination Killer: This is the most important part. Every memory fragment is hashed into a Merkle Chain. When the LLM retrieves context, the system cryptographically verifies the retrieval against the immutable ledger.

If the hash doesn't match the chain: The retrieval is rejected.

Result: The AI visually cannot "make things up" about your past because it is mathematically constrained to the ledger. Local Inference: Built on top of llama.cpp server. It runs Llama-3 (or any GGUF) locally. No API keys. No data leaving your machine.

Features:

Zero-Dependency: Runs on Windows/Linux with just Python and a GPU (or CPU).

Visual Interface: Includes a Streamlit-based "Cognitive Interface" to visualize memory states. Open Source: MIT License. This is an attempt to give "Agency" back to the user. I believe that if we want AGI, it needs to be owned by us, not rented via an API.

Repository: https://github.com/merchantmoh-debug/Remember-Me-AI

I’d love to hear your feedback on the Merkle-verification approach. Does constraining the context window effectively solve the "trust" issue for you?

It's fully working - Fully tested. If you tried to Git Clone before without luck - As this is not my first Show HN on this - Feel free to try again.

To everyone who HATES AI slop; Greedy corporations and having their private data stuck on cloud servers.

You're welcome.

Cheers, Mohamad

The Cost of AI Art

Claude created the First AI-planned drive executed another planet

Show HN: A Postgres-first platform (opening public beta)

Job Interview Questions for Embedded Systems Developers

Show HN: Synatra – OpenClaw for Teams

New coffee chemicals show promise for managing type 2 diabetes

Apple tops Q1 earnings estimates on record-breaking iPhone sales

Show HN: HN Comment Thread Analysis

Ask HN: Is OpenAI double-charging you lately?

Design, Just Rebrand Please

TrumpRx Pharmacy Program Delayed Amid Potential Anti-Kickback Concerns

The companies helping governments hack citizens' phones (2024)

Claude Code Psychosis

On Political Power

Acadia – Long-Horizon Tutoring

Breaking the QR Limit: The Discovery of a Serverless WebRTC Protocol – Magarcia

Amazon and Google Eat into Nvidia's A.I. Chip Supremacy

Show HN: A Local OS for LLMs. MIT License. Zero Hallucinations. Infinite Memory

Docker Sandboxes

Preventing Long Covid with Metformin (2026)

One-Third of Video Game Workers Laid Off in Past 2 Years

Antirender: remove the glossy shine on architectural renderings

Google DeepMind Released World Building Engine: Project Genie [video]

Building Clawdbot: Engineering Personal AI

Secure Coding Peitition for the Government of Canada

Prefix Sum on WebGPU: From Hillis–Steele, Blelloch, to Subgroups

Agent-session-commit: Update AGENTS.md at the end of a session

Actress Catherine O'Hara dies at 71

See how many words you have written in Hacker News comments

Supporting Internet-Critical Open Source Projects

Show HN: WebGPU vs. CPU Benchmark – see how much faster WebGPU is for you