How EnviroAutomate is built,
hosted, secured, and operated

Technology and Architecture

How EnviroAutomate is built, hosted, secured, and operated

EnviroAutomate is purpose-built environmental software, not a general-purpose tool with a skin. This page summarises how the platform is built, where your data lives, how we protect it, and how we keep it running. If you need a single artefact for your vendor file, the full Technology FAQ is available on request.

Where your data lives

All client data is hosted on Amazon Web Services in the Asia Pacific (Sydney) region (ap-southeast-2). Your data remains in Australian AWS data centres. We do not replicate or process client data outside Australia.

How the platform is built

EnviroAutomate runs on enterprise-grade managed cloud infrastructure:

Cloud platform: Amazon Web Services in Australia
Application framework: ASP.NET Core
Database: managed PostgreSQL on AWS
Runtime: a containerised application on serverless containers
Storage and messaging: managed object storage and managed queues for asynchronous processing
Email: transactional email through a managed email service

We name product categories on this page rather than specific service SKUs. The full Technology FAQ (available on request) lists the named AWS services and configuration detail that procurement and security reviewers typically want for a vendor file.

High availability and fault tolerance

EnviroAutomate is designed to keep running through component failures without operator intervention:

The application and database are deployed across multiple AWS availability zones with automatic failover
A managed application load balancer distributes traffic across healthy instances
Failed containers are automatically replaced
The managed database automatically fails over to a standby replica
Blue/green deployment keeps users uninterrupted during releases, with automatic rollback if a release fails health checks

Disaster recovery and backups

Recovery Point Objective in production: 5 minutes
Recovery Time Objective in production: 60 minutes
Automated database backups run daily and are retained for 35 days, with point-in-time recovery at any point within that 35-day window
Continuous transaction log backups run every 5 minutes
File storage versioning is enabled, giving full history of file changes
All backups are encrypted at rest

Encryption

In transit, all connections use HTTPS with TLS 1.2 or higher as the minimum accepted protocol
HTTP is automatically redirected and HTTP Strict Transport Security (HSTS) is enforced at the load balancer
At rest, database storage, file storage, and message queues are encrypted with AES-256
Encryption keys are managed by AWS Key Management Service
EnviroAutomate does not hold or rotate keys directly

Access, identity, and tenant isolation

Multi-factor authentication is enforced for all platform users
No user can sign in without a second factor
Single Sign-On via SAML 2.0 integrates with enterprise identity providers (Microsoft Entra ID, Okta, Google Workspace, and other SAML-compatible systems), letting your users access EnviroAutomate with corporate credentials
Tenant isolation is enforced at the application and data layer
Every query is automatically scoped to the requesting tenant, so one client's projects, sites, and analytical results are never visible to another
Tenant isolation is validated by a dedicated suite of automated tests that must pass before any code is deployed
Least-privilege access governs every internal role, both inside the application and inside our AWS environment
Production database deletion protection is enabled

Application security

A managed web application firewall with OWASP-aligned rule sets covers SQL injection, cross-site scripting, path traversal, IP reputation filtering, known-bad-input blocking, and rate limiting
Continuous threat detection runs across our AWS accounts, monitoring for unauthorised access attempts, compromised credentials, and anomalous activity
The application runs in private subnets with no direct internet exposure, and network security groups restrict traffic between components
Sensitive credentials are stored in a dedicated managed secrets service, never in source code or configuration files

Service availability

We target 99.5% uptime for production environments. We do not currently offer financial service credits. What we do provide is comprehensive monitoring, automated alerting, documented incident response procedures, and an architecture designed for redundancy and automatic failover. If we ever miss the target over a sustained period, we will tell you directly rather than discover it through an SLA dispute.

Monitoring, logging, and incident detection

Application and infrastructure monitoring tracks performance, error rates, query health, and queue depth, with automatic alarms when thresholds are exceeded
Distributed tracing provides end-to-end visibility of requests across services
Application and security logs are retained for 1 year, supporting compliance reviews and incident investigation
Web application firewall logging captures every inspected request for security analysis
Automated alerts route to the operations team immediately when issues are detected, and critical exceptions are escalated by email
Dead-letter queues capture failed background jobs for human review rather than silently dropping them

Software quality and release engineering

Every code change is automatically built and tested through continuous integration; if any test fails, the build is rejected and deployment is blocked
Our continuously growing automated test suite, currently over 9,000 tests, covers data integrity, tenant isolation, security and authorisation, scientific calculations, import and export pathways, and full report-generation approval tests where generated Word output is compared character-by-character against pre-approved baselines
Database-level tests run against a real PostgreSQL instance inside Docker containers that mirror production, catching issues that simplified test databases would miss
We typically deploy small batches of changes multiple times per week using blue/green deployment, with automatic rollback if health checks fail

AI features

EnviroAutomate includes three categories of AI-powered features designed to reduce repetitive work for environmental consultants. All three operate on data already in EnviroAutomate, are tenant-scoped, and are designed for human review before any AI-assisted output is used in a final deliverable.

The AI Assistant is a conversational assistant embedded in the platform; it can look up your sites, projects, fieldwork, and analytical data, and propose actions on your behalf, and very action that writes data requires explicit human approval
AI Assessment Review provides automated review of report types, flagging potential issues and producing structured review outputs to reduce manual checking time
AI Report Generation is AI-assisted drafting of sections within client reports, based on your project data and investigation findings; it is currently live for Preliminary Site Investigation (PSI) and Data Usability Review (DUR), with more report types in the roadmap

All AI processing occurs within Australian AWS regions. Customer inputs and outputs are not used to train foundation models.

For details about our human-in-the-loop and approval-gated actions design, see our AI safety posture. For data residency, model provider details, and AI-specific controls, see AI data and residency.

Data integration

Excel imports and exports use standardised templates
REST APIs handle programmatic data submission and retrieval
Webhooks provide real-time event notification
We support ESDAT data imports
Bespoke migration tools are developed where required to bring historical data into EnviroAutomate (e.g. where historical data is in Excel files)

Compliance posture

EnviroAutomate's security controls are designed to ISO 27001 standards
We rely on AWS's underlying ISO 27001, SOC 2, and PCI DSS certifications for the infrastructure layer
Formal EnviroAutomate certification is a roadmap item
We comply with the Australian Privacy Act 1988 and the Notifiable Data Breach scheme

For our full security posture, including subprocessors, encryption detail, incident response, and vulnerability disclosure, see the Trust overview page.

Want the detail?

Procurement and security reviewers can request the full Technology FAQ, which names specific AWS services, configuration detail, and the long-form answers behind the summary above.