Chiran Hewawitharana

"Evenza" Smart Event Management Platform is an online event management and resourcing platform that serves the functionality of an event manager as well as providing access to event-related resources by connecting relevant service providers to the customers.

The core uniqueness is that the proposed system provides a solution for planning and management of an event as well as finding necessary resources(services and products) within the same platform. The project also provides most of the basic functionality required for an event. It consists of the current background of the problems we face while planning for upcoming events and how planners can tackle these matters through this proposed system. It also entails building a fully functional system that would help achieve a well-organized, timely and within the budget event.

• Technology Stack: MongoDB, NodeJs, Express, Angular2, Swagger, Tensorflow JS, ChartJs, JWT Authentication, Socket.io, Google Cloud Platform


• Timestamp: January 2020 - August 2020
• Category: Web Development

BestDriver is an interactive web-based platform for instant and secure Tour and taxi booking mainly targeted at Foreigners traveling to Sri Lanka.

The platform is able to provide real-time taxi and tour bookings, airport pickups with a live payment calculator with the support of Google Maps, Geocoding, Matrix, and Places API. Project was developed for a client at AccSeed Ltd, where I was a co-founder since May 2019. The project was based on HTML, JS, and Bootstrap with a PHP backend.

• Technology Stack: HTML, CSS, JS, Bootstrap, PHP | Google Maps, Geocoding, Places, and Matrix APIs


• Timestamp: November 2019 - March 2020
• Category: Web Development

Pharmasiya.com was a social service platform developed by me at AccSeedd. This platform was developed to support essential service during the Covid-19 pandemic period. The platform is capable to ordering medicine online, connecting pharmacies with home-delivery. The service was provided free of charge. We were able to join with pharamacies in Western Provience and provide an outstanding service to help the community.

Our commitment has been recognized by a local News Paper, CeylonToday:: Read the Article..

• Technology Stack: HTML, CSS, JS, Bootstrap, PHP


• Timestamp: March 2020 - April 2020
• Category: Web Development

LKMarts.com online retail web stores were developed in March 2020 where the whole country was locked-down due to the covid-19 pandemic. People were suffering from the situation as they were having difficulties reaching pharmacies, groceries to fulfill their day-to-day needs We at AccSeedd identified this requirement; for an online solution to easily reach groceries to order daily products, and developed these platforms to provide the required technical support during the pandemic.

The platform was capable of providing online products ordering, SMS alerts, and a lot more features. This service was delivered free of charge as a social service. We were able to partner with more than 11 groceries in the Western Province and support the community.

Our commitment has been recognized by a local News Paper, CeylonToday:: Read the Article..

• Technology Stack: HTML, CSS, JS, Bootstrap, PHP


• Timestamp: March 2020 - April 2020
• Category: Web Development

Fully functional e-commerce web-store with real-time payments processing, user authentication, cart-add, and checkout features, build with React JS and using Firebase real-time database and Firebase cloud-functions as the backend.

• Technology Stack: ReactJs, Context-API, React Hooks, Firebase Real-time Database, Cloud Functions


• Timestamp: December 2020
• Category: Web Development

This is an absolute Slack clone, build with React JS including User authentication, Material UI for a Responsive Design, and Reach Context API(REDUX) with Firebase Authentication (Google Authentication), Firebase Cloud Firestore real-time database, and Firebase backend for real-time chat.

• Technology Stack: ReactJs, Context-API, React Hooks, Firebase Real-time Database, Cloud Functions


• Timestamp: November 2020
• Category: Web Development

This is a Netflix Clone developed using React JS, consuming TMDB (The Movie database) Public API. Database and hosting developed and enabled with Firebase. This project was bootstrapped with Create React App.

• Technology Stack: ReactJs, Context-API, React Hooks


• Timestamp: October 2020
• Category: Web Development

This is an exact Linkedin Clone developed using React JS, and Firebase as backend. Database and hosting developed and enabled with Firebase real-time database and cloud functions. This project frontend was bootstrapped with ReactJs, using Hooks and Redux.

• Technology Stack: ReactJs, React Hooks, Redux, Firebase Real-time Database, Cloud Functions


• Timestamp: January 2021
• Category: Web Development

This project is developed with python Django/Flask micro-services communicating through a RabbitMQ Event Bus. Frontend was bootstrapped with Create React App. This contains four services, built on Django and Flask, communicating through rabbitMQ. Each service is responsible for managing different activities in EazyMart Store Admin portal. Services use different min-databases implemented with MySQL and MongoDB.

• Technology Stack: Django, Python Flask, RabbitMQ, mySQL, mongoDB, ReactJs


• Timestamp: January 2021
• Category: Web Development (Microservices)

This project is the Progressive Web Application(PWA) developed for the Evenza - Smart Event Management and Resourcing Platform. Progressive Web Application generated with angular-PWA, It has the capability of managing a cache in local storage.

• Technology Stack: Angular2, NodeJS, Express, MongoDB, Angular PWA


• Timestamp: June 2020 - August 2020
• Category: Mobile Apps

DecoAR is an Augmented Reality supported mobile application which provides an e-commerce platform for interior designs and bath ware products with a realistic experience. Customers can view products in the real-world environment before purchase, which enables more convenience and reduces product returns while increasing sales dramatically with the ace of a cutting-edge Augmented Reality based experience.

The Idea was first pitched at iHack 2018 organized by University of colombo School of Computing and became the Second Runners Up in October 2018. Development was carried out until January 2020, and Deco AR won many awards in hackathons and competition.

• Technology Stack: Unity 3D, Andriod SDK, Vuforia Augmented Reality SDK, C#


• Timestamp: October 2018 - January 2020
• Category: Mobile Apps

smartCop is a smart solution based on deep learning, to manage road traffic and prevent road-accidents in an innovative; automated way. The project was developed using python, with a pre-trained Convolutional Neural Network (CNN) on Tensorflow, Keras with image-processing techniques using OpenCV.

smartCop won the Best Innovation Category Award at the Invention Convention for Engineering Smart Systems for Industry 2019 (InCo V) competition, organized by the University of Kelaniya in 2019. It also became Finalists at Codefest AI Innovator 2019 organized by the Sri Lankan Institute of Information Technology and IEEE Innovation Nation Sri Lanka Startup Pre-Accelerator in 2019.

• Technology Stack: Python, OpenCV, Google MobileNet, CNN with TensorFlow and Keras, Arduino


• Timestamp: July 2019 - January 2020
• Category: Deep Learning (CNN)

I have conducted the research project titled "A Context-aware Hybrid Personalized Recommender System for Telecommunication Service Providers in Sri Lanka." which proposed a context-aware personalized hybrid recommendation system that combined advancements in Wide and Deep learning, Factorization Machines, and user-similarity-based collaborative filtering models to recommend telecommunication services and products to achieve subscribers’ retention. Furthermore, the practical outcome of this research project resulted in the development of an enterprise product recommendation model for Sri Lanka Telecom PLC, the national telecommunication service provider in Sri Lanka. The success of this project led to its publication in the conference proceedings on Artificial Intelligence of and for Business associated with JSAI International Symposia in 2021 (IsAI-2021).

• Technology Stack: Python, Tensorflow, Pytorch, Pandas, Scipy, Azure Machine Learning Studio, SQL, Deep RNN


• Timestamp: February 2021 - January 2022
• Category: Deep Learning

This project is based on analysis and classification of news using an LSTM (Long Short Term Memory) - Recurrent Neural Network to Identify fake news over a text-based news stream. It was developed using Python, Tensorflow, pandas with spacy, Wordcloud, and gensim for Natural Language Processing and seaborn, plotly for visualizations.

• Technology Stack: Python, NLTK, Plotty, Genism, LSTM with Tensorflow and Keras


• Timestamp: October 2020 - November 2020
• Category: Deep Learning (LSTM)

This project is based on visualizing data of a sentiment analysis of tweets about US Airlines on an Interactive Streamlit Dashboards using Python. For the analysis it use classification techniques provided in NLTK Library. Plotty and WordClod visualizations are included. Streamlit is an open-source Python library that makes it easy to create and share beautiful, custom web apps for machine learning and data science. In just a few minutes you can build and deploy powerful data apps. This project generates it's visualizations using a streamlit dashboard developed.

• Technology Stack: Python, NLTK, Plotty, Streamlit, Pandas


• Timestamp: November 2020
• Category: Data Science

Accurate face landmarking and facial feature detection are important operations that have an impact on subsequent tasks focused on the face, such as coding, face recognition, expression and/or gesture understanding, gaze detection, animation, face tracking, etc. We define a face landmark as a prominent feature that can play a discriminative role or can serve as anchor points on a face graph

Face Recognition was based on two approaches using face landmarks techniques; where a Machine Learning approach with K-Means clustering to classify faces considering the Euclidean distance; and a Deep Learning approach based on a Convolutional Neural Network on TensorFlow with Keras to achieve best performance and accuracy.

• Technology Stack: Python, PyQt5, OpenCV, TensorFlow, Keras, Firebase


• Timestamp: November 2020
• Category: AI & Machine Learning

Recommendation systems can be defined as software applications that draw out and learn from data such as preferences, their actions (clicks, for example), browsing history, and generated recommendations, which are products that the system determines are appealing to the user in the immediate future.

This project contains a movie recommendation system that provides recommendations using the ALS algorithm from Spark MLLib by training a MatrixFactorizationModel.

• Technology Stack: Scala, Apache Spark, Spark MLLib


• Timestamp: October 2020
• Category: Data Science