GitHub Repository

Riverse — A personal AI agent designed for your own devices. Persistent memory, offline cognition, grows with every conversation. River Algorithm. Local-first.

11 starsPython

Riverse – Local AI agent with memory that grows over time

Name: Riverse – Local AI agent with memory that grows over time
Availability: InStock
Author: collenjk

by collenjk·Feb 25, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●●SolidBig BrainShip ItSolve My Problem

Local AI memory that actually learns, but the River Algorithm metaphor needs proof it works better.

Strengths

•River Algorithm concept is clever — sedimentation metaphor for confidence-weighted memories is non-obvious.
•Offline sleep/consolidation process mimics human cognition, genuine technical angle.
•Multi-modal input (text, voice, image) + MCP protocol + pluggable tools framework shows ambition.

Weaknesses

•No live demo or benchmarks showing memory retention/recall actually outperforms flat ChatGPT history.
•Competing against Continue, Cursor, local Ollama stacks — memory persistence alone isn't enough differentiation.

Post Description

Hey HN, I built a personal AI agent that runs locally and remembers you across conversations.

The problem: ChatGPT/Claude memory is basically a flat list — a few facts, no timeline, no confidence levels, everything in the cloud. Switch platforms and you start over.

Riverse uses what I call the River Algorithm — conversations flow through like water, important stuff settles like sediment into your profile, contradictions get washed away over time. There's an offline "sleep" process that consolidates memories, kind of like how human sleep works.

v1.0 supports text/voice/image input, Telegram & Discord bots, pluggable tools, custom YAML skills, and MCP protocol. Everything stays on your machine.

I also built a companion tool (RiverHistory) that imports your existing ChatGPT/Claude/Gemini chat history and extracts a starter profile — so your AI knows you from day one.

Would love feedback. GitHub: https://github.com/wangjiake/JKRiver