From a small startup team to a province-wide digital backbone, Pakistani tech company DAT is now powering Chief Minister Maryam Nawaz Sharif’s Suthra Punjab initiative, managing logistics for 11 Waste Management Companies across Punjab.
The company scaled its operations across the province after making a major infrastructure decision—replacing Apache Kafka with the high-performance message broker LavinMQ, enabling faster, lighter, and more reliable systems for large-scale public service operations.
Built for Low-Connectivity Environments
Originally, DAT was developed to solve workforce communication challenges in areas with low digital literacy. The platform allows field workers to communicate using familiar tools such as WhatsApp, including voice notes and Roman Urdu, making digital reporting more accessible.
As the system expanded beyond its initial scope, however, the company’s infrastructure began to struggle under increasing demand.
Scaling Challenges with Apache Kafka
DAT initially relied on Apache Kafka as its core messaging system. While powerful, it soon became a bottleneck for the startup’s fast-growing public-sector operations.
According to Ahsan Nabi Dar, the system faced recurring technical issues.
“We encountered persistent issues such as broker not found, leader missing, and duplicate messages,” he said. “For a startup trying to scale rapidly, these challenges became unsustainable.”
Performance Breakthrough with LavinMQ
Seeking a more efficient alternative, the DAT team adopted LavinMQ, a lightweight message broker known for low latency and minimal resource usage.
The results were immediate and significant. During internal testing, memory usage dropped dramatically—from around 5 GB to under 40 MB.
After migrating to the managed LavinMQ service provided by CloudAMQP, DAT reported major performance improvements:
- Daily event handling increased from 50,000 to 500,000
- Message publishing speed improved by nearly 300 times
- Operations expanded from Lahore to all 11 Waste Management Companies in Punjab
- The system was later adopted by the Punjab Cattle Market Management & Development Company and is being onboarded by the Directorate General of Archaeology, Punjab
AI System Reduces Staff Ghosting
With a more stable infrastructure in place, DAT introduced an AI-powered anomaly detection system for the Lahore Waste Management Company.
The impact was significant—field staff ghosting reportedly dropped from 14% to less than 1%, greatly improving operational transparency and accountability.
“Heart of the System”
Dar described LavinMQ as the backbone of their operations.
“LavinMQ has become the heartbeat of our system,” he said. “It manages everything from message routing and backups to critical HR processes like payroll across a province-wide workforce.”
A Model for Scalable Public-Tech Solutions
DAT’s journey highlights how strategic technology choices can determine a startup’s ability to scale. By moving from Kafka to LavinMQ, the company successfully transitioned from a small startup to a province-wide infrastructure provider supporting major public-sector initiatives.
The case is being seen as an example of how efficient, modern messaging systems can enable large-scale digital transformation in government services.
