BioGraph: environmental audit software for Form V reporting

Overview

Under the Environment (Protection) Rules, 1986, industries that operate with pollution-control consent submit an annual environmental statement in Form V to their State Pollution Control Board.

BioGraph replaces the manual preparation of that statement with a structured workflow: a company is configured by its pollution category, audit data is captured section by section, and the system generates the documents needed for submission.

The problem

Preparing Form V by hand means re-entering the same operational data — water, air, waste and energy figures — across sections, then formatting it into submission documents. Review cycles were slow, and the resulting documents were inconsistent from one submission to the next.

What I built

• Company management with category-based configuration for Red, Orange, Green and White industries, following the Ministry of Environment, Forest and Climate Change classification.
• Data models and backend flows for audit sections covering water, air, waste and energy data.
• JWT-based authentication, authorization layers and audit logs.
• A multi-format export system that generates PDF, DOCX, Excel and CSV documents from templates for regulatory submission.
• Redis caching for repeated reads, and paginated, indexed queries for large datasets.
• Dashboard interactions and implementation handoff documentation.

Architecture

The React client (built with Vite) talks to an Express API. PostgreSQL, accessed through Prisma, stores companies, audit sections and generated reports; Redis caches frequent reads and backs API rate limiting. Documents are generated server-side with dedicated PDF, Word and Excel libraries.

Engineering decisions

• Deterministic exports first. Regulatory documents have to be reproducible, so every export is generated from a template and validated data. Automation is layered on only once that output is predictable.
• Category as configuration. A company’s pollution category is part of its configuration, so the sections and requirements that apply follow from the data.
• Auditability built in. Authentication, authorization and audit logs record who changed what — a requirement for any compliance record.

Evidence and status

BioGraph is in production. The codebase is private to Ecologic Engineering, so there is no public repository; it carries automated unit, integration and end-to-end (Playwright) tests.

I’m happy to walk through the data model, export pipeline and API design in an interview.