Start a conversation
Pipelines to production

Data engineering that makes period-end boring.

The systems of record are already in your tenant. The job is wiring them into a unified operational model — one that finance, ops, and the audit team can all read from. Smelter is the engine; we’re the team that designs the schema, builds the connectors, writes the reconciliations, and runs period-end without a war room. Air-gapped where the data demands it. Lineage end-to-end.

What we deliver

What we deliver

  • Lakehouse and warehouse design (Databricks, Snowflake)
  • Spark and Airflow pipelines
  • Self-hosted ETL on Smelter
  • Master data management and lineage
  • Power BI and Looker semantic layers
  • Compliance pipelines for finance and tax
Stack we deploy

Tools and infrastructure we ship on.

Partner platforms
  • Databricks
  • Apache Spark
  • Airflow
  • Power BI
Built on engineo
  • Smelter
  • Quarry
Lakehouse
Modern data architecture
Air-gapped
Self-hosted deployments
Period-end
Compliance pipelines built in
How we engage

Pipelines designed for the runbook, not the demo.

Period-end isn't a place for surprises. We design data pipelines that hold up when the calendar is unforgiving.

  1. 01

    Source mapping

    Catalogue every system that feeds the model — ERPs, CRMs, supplier portals, third-party APIs. Define ownership, refresh cadence, and the SLAs each source can support.

  2. 02

    Model design

    Design the canonical operational model — facts, dimensions, lineage. Pick the warehouse or lakehouse architecture that matches volume, governance, and access patterns.

  3. 03

    Pipeline build

    Build on Smelter (or the customer's preferred stack) with retries, idempotency, and reconciliation baked in. Every transform is testable, every run is replayable.

  4. 04

    Operate & evolve

    Managed run with on-call ownership. New sources, new dimensions, and new regulatory requirements absorbed without breaking the pipelines that already work.

Sector scenarios

What we tend to build.

Common pipeline scopes by industry.

Finance consolidation

Multi-entity financial consolidation pipeline — automated currency translation, intercompany eliminations, and period-end controls feeding the close.

Close cycle compressed by days.
Supply-chain visibility

Real-time visibility across suppliers, warehouses, and orders — single operational view with alerting on the events that matter to ops.

Single operational view across the chain.
Audit-ready advisory

Audit-grade pipelines for professional-services firms — case data, billing, time, and partner attestation, all with lineage.

Audit-ready data, end-to-end.
Selected work

Data & ETL engagements.

Anonymised — sector and outcomes are real.

Tax & Regulatory
A diversified manufacturer
Heavy engineering & manufacturing

A diversified manufacturer ran three SKUs of business across eight plants, each generating thousands of invoices per day. GST e-invoicing and e-way bill compliance was being held together by a controller’s team spending two-thirds of every month on reconciliation. The penalty exposure was real; the burnout was visible; the existing tax tooling had not been wired correctly to the ERP.

  • Real-time IRN + e-way bill generation across 8 plants
  • Manual reconciliation hours reduced by 80%+
Data Engineering
An advisory and assurance firm
Professional services

An advisory and assurance firm needed data pipelines for client onboarding, case management, and partner billing — pipelines that would have to satisfy their own audit standards. The systems already in place were not built with lineage as a first-class concern, and the partner team had begun to lose confidence in the numbers their own dashboards were producing.

  • Audit-ready pipelines with full lineage
  • Onboarding cycle reduced significantly
Start a conversation

Bring us your hardest build.

30-minute discovery call. No decks, no boilerplate — senior engineers in the room, the same ones who'd own delivery.

Book a discovery callSee the work