Knowledge Operating System
The knowledge operating system for the enterprise.
Every organization drowns in data and starves for knowledge. Syrtis captures it from any source, keeps it live and sovereign, and delivers it on demand — every answer traceable back to its source.
The problem
Every agent, copilot and automation you stack on top is only as good as the knowledge beneath it. Give a brilliant model stale, contradictory or ungoverned knowledge and you don't get “I don't know” — you get a confident, plausible, wrong answer.
Hallucinations are almost always knowledge-layer failures: the right source was missing, out of date, or never trustworthy. No amount of prompt-engineering fixes a weak foundation.
The substrate — the context engine
Capture, index across multiple memories, then retrieve exactly what's needed — from each one — to build an agent's context. Live.
Retrieval — multi-RAG & consensus
Vector, graph, full-text, SQL: every memory is queried in parallel, then the results are crossed. What several methods confirm rises; the noise drops. Every dial is in your hands — or automated for performance.
Consensus across methods is what makes precision. The threshold tunes vector recall — raise it to cut noise, lower it to widen.
The KOS, put to work
The substrate isn't just storage. Above it, agents read the context they need and write fresh knowledge back — for concrete goals, all backed by the same foundation.
The control surface
You stay in control of every AI decision — which model, which knowledge, which database, how it's retrieved, monitored, versioned. Pick a node's model and watch sovereignty react.
Knowledge stays in the sovereign substrate; only the processing location changes.
Point sensitive steps at a local model — data never leaves your perimeter — and reserve hosted frontier models for the steps that deserve them, in the same flow.
Describe the flow in plain language; Syrtis turns it into a buildable Scenario blueprint, constrained to the real vocabulary and the 31 real nodes. An AI service in hours, not months.
Every slot and the API are versioned; the Trace node consolidates a run's full history. Quality, drift, cost, CO₂ — observable, audit-ready.
How it's built
Three applications, a message bus, a multi-store data layer, and a deterministic DAG executor. One vocabulary, from the UI down to the tables.
The graph-first scenario editor, the test chat/console, the request waterfalls and the supervision dashboards.
The single source of truth. Persists every entity, enforces access rights, records requests and bridges to Core over the bus.
The brain. Consumes jobs, runs the node DAG, drives the LLM interactions and rendering — and emits the Messages.
The library — 31 atomic nodes
Why Syrtis
The alternatives are either black boxes you can't open, or tools designed for something else and adapted to AI after the fact.
General models behind a fixed product. Syrtis opens the box: a model per step, answers grounded in your knowledge, a full execution trace.
Born for trigger→action ETL, where AI is a guest. At Syrtis, model execution and supervision are the organizing principle of the whole architecture.
Syrtis shares the open-primitives lineage — and adds everything above it: entity, convergent executor, multi-store, versioning, monitoring, access.
| Principle | Syrtis | Black boxes | Automation | Frameworks |
|---|---|---|---|---|
| Lineage | AI-native control plane | Consumer SaaS product | ETL / trigger→action | LLM libraries / GUIs |
| Agent model | Multi-agent + blackboard | None | Linear, agents bolted on late | Build it yourself |
| Knowledge / RAG | Native, tunable nodes | Hidden / limited | External wiring | Primitives, you operate them |
| Data layer | Relational · vector · graph · document | None | Connectors, no store | You provision it |
| Observability / audit | Waterfalls · lineage · Trace | None | Non AI-aware logs | DIY / add-on |
| Sovereignty / EU AI Act | By design | Depends on the vendor | Not a goal | Your responsibility |
Three years holding one direction
First end-to-end platform. Relational + vector + graph foundations. Multi-model from day one.
First usable Manager. Scenario / Card / Job / Node blocks. Parallelism. First knowledge flow.
Multi-path scenarios, conditions and loops, controlled determinism. Tunable chunking / embedding.
Graph-first editor, client SDKs, Registry, major execution upgrades, operational intelligence, multi-agent.
Out of prototype purgatory