Case Studies
Complex software. Real-world results. Long-term partnerships.
Two real projects built by 3S: an enterprise platform that grew from a small application, and a water and gas meter reading platform, enhanced by AI, running in production. Both remain in continuous development.
Case Study 01 · Enterprise Engineering Management Platform
From a small application to a business-critical enterprise platform.
How a long-term partnership turned a single project into the platform that connects a fire safety engineering company's commercial, technical, financial, and field operations.
- Industry
- Fire safety engineering & technical services
- Customer
- Confidential Brazilian engineering company
- Relationship
- Approx. 12 years, ongoing
- Delivery
- Web platform + tablet app (Flutter)
The challenge
A fire safety engineering company serving large organizations needed custom software to improve and connect its operations. The relationship began with a relatively small project — and grew as each new need was addressed.
The relationship
- Start: a relatively small project.
- Growth: gradual expansion into a comprehensive enterprise platform.
- ~5 years ago: the customer acquired ownership of the source code and kept 3S as its development partner.
- Today: 3S continues to enhance the platform.
The solution — one platform across the business
Mobile field inspections
A cross-platform tablet application for Android and iOS, built with Flutter and Dart. It works offline and synchronizes collected data when connectivity is restored.
- Locate. Technicians select fire safety infrastructure points on interactive property floor plans.
- Record. Findings are recorded on site, at each selected point.
- Evidence. Photographic evidence is attached directly to each record.
- Sync & monitor. Data syncs when online; managers and customers follow inspection progress.
Business impact
Technicians took photographs and notes during inspections, then prepared their reports after returning to the office.
Inspections are documented on site, reducing administrative work and improving productivity and report quality.
According to customer feedback relayed to 3S, the solution became an important competitive differentiator and helped the customer win new contracts.
Screenshots use anonymized data. Customer identity remains confidential.
Download Case Study 01 PDF · 1 page · 1.0 MBCase Study 02 · Hydrix: AI-Powered Meter Reading
From fragmented field processes to an AI-powered meter reading platform.
Developed and continuously enhanced over approximately 15 years, Hydrix supports water and gas meter-reading operations, processing approximately 500,000 readings per month, including 50,000 AI-assisted readings. It began as a spreadsheet automation project and grew to add mobile field collection, AI reading extraction, and human review.
- Industry
- Water & gas meter reading services
- Experience
- Approx. 15 years
- Operational volume
- Approx. 500,000 readings / month, incl. approx. 50,000 AI-assisted
- Historical expansion
- Original customer + 5 additional reading companies
The challenge
Meter reading companies collected readings in the field and submitted them to property management companies — each of which required the data in its own spreadsheet format.
How the solution evolved
- Stage 1
Automated spreadsheet generation and distribution in customer-specific formats.
- Stage 2
Expansion from the original customer to five additional meter reading companies.
- Stage 3
Mobile app for field photos and readings, with offline operation and later sync.
- Stage 4
AI-powered extraction of readings directly from meter photographs — for water meters and, since October 2026, gas meters.
AI with validation built in
Image recognition is combined with historical readings, business rules, and exception handling. Uncertain or inconsistent readings can be routed for human review.
Current capabilities
- Mobile field collection
- Offline synchronization
- AI-powered reading extraction
- Historical validation
- Exception handling
- Operational dashboards
- Reporting
- Automated file export & distribution
Included in the total: approx. 10% of current monthly volume.
Annualized: ~6 million total and ~600,000 AI-assisted readings / year. Current monthly volumes extended over twelve months; not separately measured annual or historical totals.
Screenshots use anonymized data.
Download Case Study 02 PDF · 1 page · 843 KBContact
Let's talk about your next project.
End-to-end projects, dedicated teams, individual specialists, or white-label development. Tell us what you need and we will evaluate scope and team availability with you.