ML.
Home

All projects

A collection of things I've built.

2025full-stack

Speech Therapy Bot

AI-powered speech rehabilitation platform delivered via WhatsApp

An AI-powered speech therapy platform for post-stroke aphasia patients, delivered entirely through WhatsApp. Uses Google Gemini for real-time speech analysis and personalized feedback, with an AdminJS back office for therapist management.

  • AdonisJS
  • TypeScript
  • Google Gemini
  • LangChain
  • +5
2025full-stack

Kinso

AI-powered relationship manager for founders and operators

An intelligent assistant that enriches your contacts from various sources, integrates with your calendar, and prioritizes meetings. Built as lead engineer using Expo, NestJS, and LangGraph for agentic AI workflows.

  • Expo
  • NestJS
  • LangGraph
  • PostgreSQL
  • +2
2024full-stack

QatarEnergy Privilege Program

Large-scale loyalty platform for Qatar's national energy company

Comprehensive loyalty and rewards platform serving thousands of QatarEnergy customers. Led mobile, web, and backend teams to deliver MVP in 3 months with scalable Azure infrastructure.

  • React Native
  • Next.js
  • NestJS
  • Azure AKS
  • +4
2023mobile

The Pekoe Trail

Hiking navigation app for Sri Lanka's premier long-distance trail

A comprehensive mobile application for The Pekoe Trail in Sri Lanka, featuring offline navigation, trail information, and community features. Delivered in 3 months using React Native and Mapbox with full offline functionality.

  • React Native
  • Mapbox
  • Offline Maps
  • React Navigation
  • +1
2023mobile

Swingzen

AI-powered golf swing analysis application

An innovative golf coaching tool that uses AI and computer vision to analyze golf swings in real-time. Integrated machine learning models with React Native Vision Camera for frame-by-frame analysis.

  • React Native
  • React Native Vision Camera
  • TensorFlow Lite
  • AWS
  • +2
2022mobile

Smart Scanner

AI-powered electricity meter reading app for Sri Lanka

Mobile application using computer vision to automatically read Sri Lankan electricity meters. Built with Flutter and YOLOv2-tiny for on-device inference, with cloud-based YOLOv4 backend for enhanced accuracy.

  • Flutter
  • YOLOv2-tiny
  • TensorFlow Lite
  • AWS Lambda
  • +2