T. Harith Withana

About Me

I’m a professional with over 4 years of work experience in the data science field, currently pursuing my Master’s in AI at RMIT Melbourne. I’m deeply passionate about AI/ML and eager to contribute to its advancement. Committed to staying abreast of emerging technologies, I’m poised to make meaningful contributions to any team or project and to expand my horizons as well as attain more qualifications from the experience I get.

Education

Master in AI | RMIT University (2024-2025)

Main Modules : Practical Data Science with Python, Big Data Processing, Social Media Analysis, AI, Programming Autonomous Robots, Cloud Computing, Computational ML, Deep Learning
Minor Thesis : “Using Large Language Models to Simulate Users to Analyse Music Recommendations” : LLM–based user-simulation together with standard recommender-system pipelines, prompt-engineering workflows, fairness-aware post-processing, and music-listening datasets such as LFM-1b and Spotify open resources.

B.Tech., Computer Science & Engineering | National Institute of Techonlogy Durgapur (2015-2019)

Graduated with a First Class.
Main Modules : Object Oriented Programming, Operating Systems, Data Structures, DBMS, Microprocessor, Machine Learning, Software Engineering, Algorithm Analysis and Design, Advance Graph Theory, Image Processing, Advanced Computer Architecture

Professional Work Experience

ML Engineer @ Integra Technology Solutions (July 2025 - Present)

Leverage and fine-tune LLMs and ML to automate document processing and insight extraction.
Build microservices for each AI feature to enhance user experience, integrating MongoDB (NoSQL) and AWS S3 for scalable data handling.

Project Manager @ Omdena (April 2024 - January 2025)

Omdena top talent opportunity
Part time role managing a team of Engineers for the project “Identifying Inaccuracies in GESI Conversations : Sri Lanka”

Machine Learning Engineer @ Intermediate Data Systems (November 2023 - February 2024)

Omdena : Top Talent Opportunity
Design Cloud archtectures for various NLP and LLM projects
Rigth Party to contact with predicting the time and day
Customer Segmentation for a Retail Mass business to give the bussiness more insights
Brainstorm with stakeholders regarding different solutions

Senior Data Scientist @ EngenuityAi (July 2021 - February 2024)

R&D projects : Satellite/LiDAR for Forestry Conservation
Research on how to track the forest cover using Satelites (only using opensource resources)
Leading a team of Data Scientists/Analysts
Directly working with stakeholders and the senior management
Applying ML modeling for real world business senarios in finance sector and IOT
Visualize data with BI tools
Computer Vision based object detection modeling for Insurance

ML Engineer @ Omdena (July 2020 - February 2024)

Omdena is the collaborative platform consists of Data Scientists from all around the world to build innovative, ethical, and efficient AI and Data Science solutions to real-world problems
Computer vision based Trash classification and tracking, Scocial meia anlysis and recomender systems

Jr Data Scientist @ LB Finance PLC (_February 2020 - July 2021)

Work with stakeholders throughout the organization to identify opportunities for leveraging company data to drive business solutions
Mine and analyse data from company databases to drive optimization and improvement of product development , marketing techniques and business strategies
Develop custom data models and algorithms to apply to data sets
Use predictive modeling to increase and optimize customer experiences, revenue generation , ad targeting and other business outcomes
Develop processes and tools(Tableau etc.) to monitor and analyse model performance and data accuracy

Projects

Optimizing Solutions for Identifying Inaccuracies in GESI Conversations in Sri Lanka 2024 April-

This is another Omdena top talent project to build a system leveraging NLP and LLMs that identify inaccuracies and misleading content in Gender Equality and Social Inclusion (GESI) conversations addressing minorities in Sri Lanka.
I perform the role in Project Lead where I guide a small team of developers

Predicting the best day,time to call clients for business 2023 Dec-2024 Feb

Contributed to an Omdena top talent project focused on developing a predictive system to determine the most effective day and time to contact clients for a major insurance company in South Africa
Led data cleaning and feature engineering processes to prepare the dataset for machine learning model development
Developed and deployed machine learning models using AWS SageMaker, leveraging advanced algorithms to predict optimal client contact timing for improved business outcomes

Remote Sensing: Carbon Tracking and Forest Monitoring Projects Link 2023 Sep - 2024 Feb

Led a cross-functional team of data scientists and collaborated with the front-end team on an R&D project focused on leveraging remote sensing technologies such as LiDAR and open-source satellite images (Optical and Radar) for forest monitoring.
Developed machine learning techniques to track forest characteristics (e.g., health, wildfire risk, soil moisture) and estimate carbon stock, contributing to ongoing improvements and research in Phase 1 of the project.
Spearheaded backend development using AWS services (EC2, S3, Lambda, API Gateway) to create API calls for accessing the models, enabling seamless integration and deployment.
Conducted extensive research in collaboration with professors from the University of Colombo School of Computing to enhance accuracy in tracking forestry and estimating carbon stocks using satellite data.
Aligned the project’s objectives with forest conservation efforts and explored opportunities in the carbon trading business for sustainable environmental practices.

Vehicle Damage Detection and Classification Using Computer Vision Project Link 2023-2024

Led a team of Data Scientists for this project involving research, data extraction, model building, evalution, back-end development and working with front-end team
Curated high-quality dataset through web scrapping and wrangling from auction sites, annotated the dataset matching with the insurance standards
Used state of the art computer vision, deep learning algorithms for vehicle classification, number plate recognition, vehicle part detection, vehicle damage detection and damage severity estimation

Sales Forecasting: Beverage Company in Sri Lanka Project Link 2023-2024

Use customer data in a JDE platform, extract data and anlyzed them.
Built ML models to forecast sales with the help of extracted data from internal and external sources.
Built PowerBI dashboards to report and visualize the outputs.

Identifying and Monitoring Human Rights Abuses Through NLP Project Link 2021 May-July

Use social media data(Twitter and Reddit) scrapped to identify any Human Rights violence happening around the world and visualize on a Dashboard
In this group project I was involved in social media data scrapping , data labeling, Classification modeling using Transformers and ML(XGBoost
Classifier), and Database development

NLP Empowered Climate Data Dashboard: Visualizing Climate Change Impacts and Mitigating Nature-Based Solutions Project Link 2020 Sept-Nov

The objective of the project was to assess coalition websites and their respective members’ websites
In this group project I was involved in web scrapping of the coalition websites, data labeling, and feature extration for Sentiment analysis and recomender systems

Trash Detection and Predicting Illegal Dump-sites Project Link 2020 June-Aug

Engaged in a dual-purpose project aimed at enhancing waste management by predicting trash types and appropriate disposal methods, as well as identifying illegal dumpsites across Europe
Involved in data acquisition efforts through web scraping, annotated datasets for machine learning, and contributed to the development of a detection model for precise trash classification and the construction of a robust back-end database
Performed comprehensive data wrangling and feature extraction to facilitate the detection of illegal dumpsites

FD Demand Prediction Creating a model for Predicting the demand for FD and build live Dashboards for a Finance Institute 2020 May-October

Spearheaded the enhancement of the data warehouse infrastructure to support Data Science (DS) and Business Intelligence (BI) functionalities, utilizing SQL Server Management Studio (SSMS) for data preprocessing and warehousing
Developed and trained a machine learning model to forecast 12-month demand across all branches, leveraging historical data and advanced modeling techniques
Designed and implemented Tableau dashboards to visualize model predictions and create interactive BI tools for real-time decision-making

Final Year Research Project: Handwritten Character Recognition Using Convolutional Neural Networks NIT Durgapur — Jan 2019 - May 2019 Supervisor: Prof. Debashis Nandi

Developed a deep learning model specializing in the segmentation and recognition of postal addresses on handwritten letters
Addressed challenges in accurate segmentation by implementing a Single Shot Multibox Detector (SSD) model using TensorFlow
Enhanced training by integrating a custom dataset provided by the Indian Postal Service, achieving significant improvements in model accuracy

Technical Skills

Programming Languages: Python, Java, R, HTML/CSS
Databases: MySQL, PostgreSQL
Machine Learning & Data Science: Machine Learning, Natural Language Processing (NLP), Deep Learning Frameworks; TensorFlow, PyTorch, Image
Processing(OpenCV), LLM
Operating Systems: Windows, Linux, Mac
Data Visualization: Tableau, PowerBI
Cloud Computing: AWS Services EC2, S3, Lambda, RDS, ECR, Sagemaker, Bedrock

Certifications

Generate Synthetic Images with DCGANs in Keras Generate Synthetic Images with DCGANs in Keras Link
Deep Learning Specialization 2020 Link
Microsoft Technology Associate for Software Development Fundamentals Link

Contact

S4078543@student.rmit.edu.au

tharinduharith.srt@gmail.com