Loading Senga…
Three layers, one system. Each is independently useful; together they replace the data plumbing every operations team rebuilds from scratch.
Point an adapter at each source. senga introspects their schemas, maps them into virtual entities you can query — without copying a row. No warehouse, no nightly ETL, no migration.
Which projects are active?
What are their rates and quantities?
Ask in plain English — or write SQL — over your virtual schema. senga's planner pulls from each source, runs the join, and streams the result. You don't care whether a column lives in Postgres, an Excel file, or an IFC model.
That unified result isn't the end — it's a starting point. Pour it into Word, PDF or PEPPOL XML using your templates; pin it to a live dashboard for KPIs and charts; or fire it into a workflow — calendar entries, planning timelines, webhooks. Same query layer, many destinations.
The spreadsheet. The ERP. The building model. Give them a common structure, and ask questions across all of them.
Your data stays where it lives.
Tables, fields and relationships.
One question, a connected answer.
Read the items, units and prices in a spreadsheet. Join them to project quantities when you need an estimate.
Choose a question and press Run query. This interactive example shows how sources combine into one result, with each row traced back to its origin.
from project ppostgresjoin budget b on b.id = p.budgetexceljoin invoice i on i.proj = p.idodoowhere p.status = 'active'having sum(i.amount) > b.cap
Review the question, then press Run query to explore the example data.
Follow the same data from its source to the work it powers: a query, a quotation, an invoice, or your next decision.
Each adapter exposes its source as virtual tables. The schema registry tracks types, primary keys, foreign keys, and source provenance. Rename a column upstream, get a typed warning before the query breaks.
senga's IFC adapter parses model files into queryable entities — elements, quantities, materials, spatial relationships. Join an IFC model against your rate book and you have a costed bill of quantities in one query.
Write one query, hit run. The planner decomposes it into per-source subqueries, pushes WHERE clauses down where the adapter supports it, and runs the joins in senga. You see EXPLAIN output, plan caching, and per-step timing — like a real database.
Show active projects with their rates and quantities.
Save any cross-source query as a view, drop it onto a dashboard as a KPI card, sparkline, or table. Each render reads fresh from the underlying systems — no extraction step, no nightly refresh. Embed in your intranet or share a link.
Built-in PEPPOL access point. Outbound BIS Billing 3.0 invoices and credit notes signed with your certificate, routed to the right participant — no separate vendor, no extra integration.
Save any cross-source query as a rule, then schedule it, watch it for a condition, or hand the result to a downstream system. Generate quotations every Monday morning. Email the project lead the moment a budget crosses 95 %. Add purchase orders to your planning calendar as soon as they're acknowledged. Same query layer, fired automatically.
senga ships as a single binary or container. Run it on-prem, in a VPC, or on European hyperscalers. No data leaves your perimeter. No credentials sit on a SaaS we operate.
Three steps from deployment to useful answers. Start with the systems you already run and build from there.
Run the senga binary or Helm chart on your own infrastructure. Point it at a Postgres for metadata, a key-store for credentials, and your identity provider. Up in an afternoon.
Configure adapters for the systems you already run. Each one introspects its source, registers virtual tables, and exposes them in the schema registry. No data is copied.
Write SQL across sources — or build queries visually. Save them as views, schedule them, render them into Word, PDF or PEPPOL XML, or hand them to your own apps via the API.