AIoT · Aquaculture · NIC Hyderabad
AquaGrid
AIoT smart aquaculture platform: sensors, computer vision, and AI insights for fish farms in Sindh, Pakistan.
Problem
Fish farmers in Sindh often lack real-time visibility into pond conditions. Decisions rely on manual observation, which means delayed reactions to dissolved oxygen, pH, temperature, and ammonia swings, driving avoidable mortality and wasted feed.
Solution
AquaGrid connects pond IoT sensors to a cloud platform with AI/analytics and a React Native app. Farmers get live water-quality monitoring, threshold alerts, farm surveillance with computer vision (YOLO), and actionable recommendations, so management is driven by actual pond data, not guesswork.
Role
Final Year Project · Founder / Full-Stack & AIoT Engineer · Incubated at NIC Hyderabad
Key features
- Real-time water-quality monitoring: DO, pH, temperature, ammonia, TDS, turbidity
- IoT pipeline: ESP32 sensors → MQTT → cloud/backend → Supabase → mobile dashboard
- Computer vision with YOLO for farm surveillance and risk signals
- Alerts when parameters cross thresholds, before losses compound
- AI layer for disease-risk hints, feeding recommendations, and predictive insights
- Multilingual AI/RAG assistant support: English, Urdu, and Sindhi
- Historical farm data and analytics for data-driven decisions
Outcome
Delivered as a final-year project with a working app and public site at aquagrid.tech. Incubated at NIC Hyderabad and presented across Teknofest, Procom Agentic AI Hackathon, SMIU/DITE, Sustainovation, YESIST12/WePOWER, and Eco Compute & Smart Cities Summit.
Stack
