Free 30-minute AI & automation audit with a senior engineer. You keep the notes. Book yours

Use case · Manufacturing & sustainability

Product-level carbon accounting and CBAM reporting for industrial exporters

A multi-tenant SaaS that maps plants, processes and product routes, calculates emissions and product footprints, and generates CBAM workbooks and ESG reports.

Client details are kept confidential. This use case describes the problem and what we built.

The brief

A climate-tech company set out to help manufacturers measure and report their carbon emissions in enough detail to satisfy regulators and customers. Its first focus was industrial producers exporting to the EU, who need product-level embedded emissions for CBAM.

We engineered the core platform with them: the multi-tenant backend services, the calculation logic for production routes and footprints, the data integration and reporting functions on Google Cloud, and the Angular application their customers use every day.

The challenge

What needed solving

Regulators now ask for product-level emissions

Exporters of carbon-intensive goods to the EU must report embedded emissions under CBAM. That requires emissions per product, not just a company-wide total.

Emissions follow the production route

Materials move through installations and processes before becoming finished goods. Footprints depend on mass balance, emission factors and how each product flows through the plant.

Data lives in many systems

Fuel, electricity, material consumption, fleet and supplier data come from ERP exports, spreadsheets and meters, in different formats and units.

Each customer's data must stay separate

As a SaaS product, the platform had to isolate every organization's data while sharing one code base.

The solution

What we built

  1. Multi-tenant GHG activity service

    A Spring Boot 3 service on PostgreSQL resolves the tenant on each request and routes it to that organization's data partition. It stores GHG activities, emission factors and carbon content, with Okta-based authentication.

  2. Installation and product-route modelling

    Installations, processes, nodes and edges form a graph of how products move through a plant. Calculation services apply mass balance and emission factors along each route to produce product footprints and intensities.

  3. Asset and supply-chain registers

    A separate asset service manages facilities, equipment, fleet, asset units, buyers and suppliers, feeding Scope 1, 2 and 3 calculations.

  4. Data integration and analytics

    Users map uploaded data to platform fields in the browser. Python Cloud Functions on Cloud SQL serve consumption data, aggregates and summaries, and BigQuery supports analytics.

  5. Automated regulatory and ESG reporting

    A Cloud Function fills the CBAM communication template as an Excel workbook from calculated data and stores it in Cloud Storage. A report service manages GHG report sections and responses and renders ESG reports to PDF.

  6. Angular application and delivery pipeline

    An Angular 17 app with PrimeNG, AG Grid and Highcharts covers onboarding, measure, reduce, inventory, process flows, data quality and reports. Azure DevOps pipelines build Docker images and push them to Google Artifact Registry.

The outcome

What changed

Footprints per product, not per company

Emissions are traced along each product's production route, which is what CBAM and customer disclosures ask for.

Reports generated, not assembled

CBAM workbooks and ESG reports come straight from calculated data instead of being compiled by hand in spreadsheets.

Ready for many organizations

Tenant isolation and per-organization data partitions let the product onboard new customers on the same platform.

Technology used

  • Java
  • Spring Boot
  • PostgreSQL
  • Flyway
  • Okta
  • MapStruct
  • Python
  • Google Cloud Functions
  • Cloud SQL
  • BigQuery
  • Google Cloud Storage
  • pandas
  • openpyxl
  • Angular
  • PrimeNG
  • AG Grid
  • Highcharts
  • Tailwind CSS
  • Docker
  • Azure DevOps
  • Google Artifact Registry

Have a similar problem to solve?

Share a few lines about your project. A senior engineer — not a salesperson — replies within one business day.

Free AI audit Let’s talk