Launching today
Sirca
Give your AI agents your company's full context
3 followers
Give your AI agents your company's full context
3 followers
Your AI agents are only as good as what they know about your company, and most of it is chunked into a vector store. RAG can't tell two people with the same name apart or know what was true last quarter. Sirca connects Drive, Slack, Salesforce, Postgres and more into a knowledge graph with real entities, relationships and history, served through one API that enforces permissions down to the embedding, cites every answer, and surfaces conflicts instead of burying them. Free to start, no card.
Hi everyone, I’m Zaki, founder of Sirca.
Sirca started from a very specific problem: making AI understand complex, high-value physical assets.
Our first major use case was Honeycomb, our asset-intelligence product for energy and resources.
In mining, oil & gas, renewables, and infrastructure, the information around an asset is not simple. You are dealing with technical reports, financial models, operational data, geological assumptions, production histories, contracts, risk registers, maps, permits, ESG data, market context, and years of fragmented decision-making.
That data is not just “documents.”
It is a living body of knowledge around an asset.
And if an LLM does not understand the relationships inside that body of knowledge, it cannot reason properly.
It might retrieve the right paragraph and still miss the point.
It might answer confidently while ignoring the financial model.
It might cite a technical report without understanding how the assumptions flow into valuation, risk, operations, or investment decisions.
That was the core problem we had to solve with Honeycomb:
How do you make AI understand the full context around complex assets, not just search across files?
To do that, we had to build the underlying technology first.
That became Sirca Platform.
Sirca is the context layer for AI inside organisations.
It connects company data, structures it into a living knowledge graph, preserves source evidence, maps relationships between entities, and gives people and AI systems a trusted layer to query.
So instead of every AI workflow needing its own fragile retrieval setup, Sirca gives the organisation a shared foundation of structured, permission-aware, evidence-backed context.
What you can do today:
• Create a workspace
• Connect or add company knowledge
• Query it in natural language
• Inspect the sources behind answers
• See relationships between documents, entities, decisions, risks, projects, and workflows
• Start building AI workflows on top of a real context layer
Honeycomb is where the technology was first built and tested.
Sirca Platform is the horizontal layer underneath it - built for any organisation that needs AI to understand complex internal knowledge before it can act intelligently.
The bet is simple:
AI apps are only as good as the context they can access.
Every serious company will need an internal context layer between its data and its AI systems.
That is what Sirca is building.
Free workspace, no card required.
We'd love your feedback, especially if you're building AI tools that need to reason over messy real-world company knowledge. Drop a comment or reach me at zaki@sirca.io.
Huge thanks to everyone checking it out today.
Zaki