what is arche?
arche is persistent memory for AI agents. it stores what AI learns about you—preferences, context, history—and delivers it to any model, any tool, any agent. one identity across every conversation.
your memory in action
see how arche remembers the details that make each conversation uniquely yours
interface preferences
prefers dark mode in all applications, uses vim keybindings, keeps terminal sessions organized with tmux
professional context
works at series b fintech startup, specializes in backend systems, leads engineering team of 8
personal habits
morning coffee routine, prefers remote work, enjoys hiking on weekends, vegan lifestyle
technical expertise
expert in python and go, uses docker for development, prefers postgresql over mysql
communication style
prefers async communication, direct feedback, values thorough documentation
learning preferences
learns best through hands-on projects, prefers video tutorials over reading, takes detailed notes
how arche works
memory that moves with you
ingest
import from chatgpt exports, linkedin profiles, resumes, or manually add memories. arche extracts and classifies the signal from noise.
supports multiple formats: chat logs, social profiles, documents, manual entry
embed
each memory is vectorized and stored as embeddings. episodic, semantic, and procedural knowledge—all searchable by meaning.
768-dimensional vectors using state-of-the-art embedding models
inject
via mcp, your ai queries your memory before responding. relevant context is injected automatically. you stay in control.
real-time semantic search with configurable similarity thresholds
self-hosted architecture
arche runs entirely on your infrastructure. your data never leaves your control.
your data stays yours
deploy on your own postgresql database with pgvector
run the mcp server on your infrastructure
export anytime, no vendor lock-in
full compliance with your data regulations
PostgreSQL + pgvector
your vector database
Model Context Protocol
your ai agents connect here
Web Dashboard
manage your memories
ready to build your memory?
join developers, researchers, and power users who use arche to maintain context across every ai interaction.