# ResDB

> Organizational memory infrastructure for the enterprise. ResDB gives organizations
> and AI agents associative memory — surfacing patterns, connections, and causal chains
> across decades of accumulated knowledge. Not retrieval. Resonance.

ResDB is infrastructure that developers and enterprises build on top of. It uses
attractor-based resonance rather than RAG-style nearest-neighbor search, activating
coherent knowledge clusters instead of returning keyword-matched fragments.

Search finds what's similar. Memory surfaces what's coherent.

## Key Concepts

- **Organizational memory infrastructure**: A new category of enterprise infrastructure that preserves and activates institutional knowledge for both human teams and AI agents.
- **Attractor-based resonance**: A query mechanism that seeds a dynamic field with a question and surfaces the stable knowledge structure that emerges — discovering patterns across departments, years, and previously unlinked information.
- **Associative memory**: The capability to recall knowledge by pattern activation rather than keyword search, mirroring how experienced humans think.

## How ResDB Differs from RAG

RAG (retrieval-augmented generation) uses nearest-neighbor search to find document chunks similar to a query. It is retrieval, not memory. ResDB activates coherent knowledge clusters, preserving context, causality, temporal relationships, and contradiction. RAG is a very fast filing cabinet. ResDB is organizational memory.

## Core Resources

- [Homepage](https://resdb.ca/): Overview of organizational memory infrastructure, capabilities, and early access
- [How It Works](https://resdb.ca/how-it-works/): Six-layer architecture explainer — hyperdimensional computing, attractor basin dynamics, temporal decay, hash-chained provenance, LNN homeostasis, and Resonant Query Language (RQL)
- [Technical Whitepaper](https://resdb.ca/whitepaper.html): Architecture details — hyperdimensional computing, resonance algorithm, temporal property graph, query language
- [Early Access Program](https://resdb.ca/early-access-info.html): Who early access is for — enterprise teams and integrators, licensing model, expectations
- [Request Early Access](https://resdb.ca/early-access.html): Beta signup form for Q2 2026 launch
- [Blog](https://resdb.ca/blog/): Long-form thinking on organizational memory infrastructure, enterprise AI, and institutional knowledge
- [News](https://resdb.ca/news/): Product milestones and company announcements

## Documentation

- [ResDB Whitepaper PDF](https://resdb.ca/ResDB-Whitepaper.pdf): Executive overview of organizational memory infrastructure

## Technical Architecture

ResDB combines Hyperdimensional Computing, typed temporal property graphs, and attractor-based resonance dynamics into a unified memory substrate. Key specifications:

- P95 query latency: ≤ 1.5 seconds
- Provenance coverage: ≥ 95%
- Hash-chained provenance on every result
- GPU-accelerated (NVIDIA CUDA, FAISS indexing)
- Native Model Context Protocol (MCP), REST API, Agent-to-Agent protocol
- Resonant Query Language (RQL): declarative, temporal, provenance-aware

## Blog

- [Search is not memory — and the difference costs enterprises millions](https://resdb.ca/blog/organizational-memory-vs-rag/): Explains why RAG-style retrieval is fundamentally different from associative memory, and why that distinction determines whether AI agents can reason or merely retrieve. (Published 2026-03-25)

## News

- [Early Access Program is now open — Enterprise and Integrator tracks](https://resdb.ca/news/early-access-program-launch/): ResDB's early access program is now accepting applications across two tracks — Enterprise (internal deployment) and Integrator (building products on ResDB). Limited to 12 organizations in the first cohort. (Published 2026-03-20)

## Company

ResDB is developed by Antturi Media Ltd. Beta available Q2 2026.

- Website: https://resdb.ca
- LinkedIn: https://www.linkedin.com/company/resdb/
- Contact: info@resdb.ca