I'm a creative and impact-driven Data Scientist and AI Engineer with 4+ years of experience in deep learning, statistical modeling, and natural language processing. My passion lies in transforming complex data into meaningful insights and building intelligent systems that solve real-world problems. I specialize in Python-based machine learning frameworks such as PyTorch and TensorFlow and bring hands-on experience with LLMs (e.g., GPT-4, LLaMA), Gaussian Processes, and Machine Learning algorithms.
I've developed solutions across various domains, such as automating spreadsheet explanations using LLMs, improving climate predictions for the Hawaiian Islands through statistical downscaling, and building anomaly detection models for large-scale retail operations. My work bridges research and application and has contributed to academic publications and technical reports.
Beyond data science, I'm passionate about teaching, research, and community-driven problem-solving. I enjoy collaborating across disciplines, mentoring students, and presenting findings to technical and non-technical audiences.
In my leisure time, I love playing chess, going to the gym, hiking, exploring new places, and meeting new people.
Let's connect and explore how data and AI can make a difference together.
I was a Graduate Teaching Assistant (TA) in the Department of Information and Computer Sciences at the University of Hawai'i at Mānoa.
During Fall 2022, I supported instruction for ICS 332: Operating Systems and ICS 451: Data Networks. In Spring 2023, I continued as a TA for ICS 332 and contributed to ICS 355: Security and Trust I. In these courses, I was responsible for grading assignments, providing detailed feedback, addressing student questions, and supporting instructional activities in the theoretical and practical components of the curriculum.
In my role, I assessed homework for over 50 undergraduate students and provided detailed feedback to clarify complex topics and reinforce core course concepts. I held office hours to answer student questions and improve their understanding of the course material. Student feedback shows a significant positive impact from my support, with 86% of respondents rating their experience with my teaching assistance as a 4 or 5 on a 5-point scale. This high level of satisfaction emphasizes the effectiveness of my instructional support and mentorship.
Also, I conducted practical hands-on demonstrations using Wireshark and Vagrant, bridging theoretical and practical knowledge. These demonstrations increased student engagement and provided a real-world context for networking and systems concepts. My teaching assistantship spanned the following courses: ICS 451 (Data Networks - Fall 2022, Fall 2024, Spring 2025), ICS 332 (Operating Systems - Fall 2022, Spring 2023), and ICS 355 (Security and Trust I - Spring 2023).
I contributed to advancing environmental modeling for the Hawaiian Islands as part of two machine-learning research projects.
In the first project, supported by the CHANGE-HI grants (NSF #OIA-2149133, NSF #2238375, and PI-CASC G21AC10381), I worked on statistical downscaling for future rainfall prediction in Hawaii. My responsibilities included generating rainfall forecasting maps by extracting CMIP6 Global Climate Models (GCMs) and applying statistical downscaling techniques using Python to support improved water resource management. I implemented site-specific linear regression models, Gaussian Processes using GPyTorch, and deep neural networks to establish baseline and advanced prediction models. These efforts contributed to a peer-reviewed journal paper (submitted - addressing revisions), where I co-authored novel AI-based methodologies and prepared results using LaTeX. I also developed rainfall map visualizations to understand spatial rainfall trends across the Hawaiian Islands.
The second project, in collaboration with MITRE, focused on estimating net radiation over the islands at a 250-meter spatial resolution and hourly temporal resolution. I predicted incoming long-wave radiation during nighttime hours using simple linear regression and more complex neural network models implemented in TensorFlow. I analyzed satellite imagery and contributed to a technical report documenting the methods and findings.
In both projects, I collaborated with computer scientists, geographers, and industry partners to integrate domain expertise with technical modeling. I also presented research findings at technical workshops, which improved stakeholder engagement and broadened the projects' visibility within academic and applied research communities.
From August 2021 to July 2022, I worked as a Lecturer on Contract in Department of Computer Engineering, Faculty of Engineering, University of Peradeniya. In this role, I was the Instructor In Charge (IIC) for several undergraduate courses, including Computer Communication Networks, Embedded Systems, and Network & Web Application Design. I led tutorial sessions, student discussion sessions, and labs, focusing on hands-on learning and the practical application of concepts using programming tools like OpenCV, Python, C, and Java.
In addition to my primary courses, I contributed to interdisciplinary teaching by covering modules in Image Processing (CO543) and Networking for Electrical Engineering (CO253), supporting the Electrical Engineering curriculum. I collaborated with senior lecturers to create assignments, exams, and lesson plans that enhanced student engagement and improved course delivery. My role balanced independent instruction and teamwork, ensuring students gained a practical understanding of computer and electrical engineering.
From August 2020 to August 2021, I was a Temporary Instructor (Graduate Teaching Assistant) in the Department of Computer Engineering, Faculty of Engineering, University of Peradeniya. During this period, I supported several undergraduate courses, including Computer Architecture, Advanced Computer Communication Networks, Embedded Systems, and a first-year computing course. I served as the Instructor in Charge (IIC) for several courses, responsible for developing and grading assignments for classes of about 60 students each semester. My tasks included preparing quizzes, designing lab exercises, and conducting review sessions to reinforce key concepts.
I automated the grading of the Network and Web Application Design course with GitHub Classroom and unit tests through GitHub Actions, making grading more efficient and consistent. This change significantly reduced grading time and sped up the feedback process. I also led a team of four teaching assistants for the Embedded Systems course, ensuring fair assessments across all lab sections while utilizing tools and programming languages like Git, GitHub, Python, C, and Java to manage coursework and support student learning. The courses I contributed to include CO513: Advanced Computer Communication Networks, CO324: Network and Web Application Design, CO321: Embedded Systems, CO224: Computer Architecture, and CO323: Computer Communication Networks II.
From February 2019 to July 2019, I worked as a Trainee Associate Software Engineer at Zone24x7 (Pvt) Ltd in Sri Jayawardenepura as part of the Big Data and Data Science team. I contributed to two major projects involving artificial intelligence solutions for KOHL, a large U.S.-based retail chain. I collaborated with my supervisor, Hansa Perera, on a log-based machine learning R&D project that aimed to predict critical events by analyzing millions of system log files. I performed detailed data analysis, implemented topic modeling to cluster log events, and extracted pattern sequences to feed into a neural network for forecasting future log events.
In a second project focused on video-based machine learning, we examined in-store video feeds to identify anomalies in customer behavior. I developed machine learning components for anomaly detection, specifically utilizing the One-Class SVM algorithm to pinpoint outliers in customer visit patterns. This effort improved the accuracy of our detections and provided valuable insights for operational decision-making.
I drove data science deliverables throughout both projects by working closely with project managers, software engineers, system architects, and QA testers. I participated in sprint meetings, presenting updates and model performance insights to the client and the CTO. These updates helped shape the client's AI adoption strategies and highlighted the value of applied machine learning.
1. A. Indika*, N. Warusamana*, E. Welikala*, and S. Deegalla. "Ensemble stock market prediction using SVM, LSTM, and Linear Regression." Authorea Preprints, 2021. DOI:
10.36227/techrxiv.16626019.v12. Y. Hatanaka*, A. Indika, T. Giambelluca, and P. Sadowski. "Statistical Downscaling from Sparse Observations with Deep Learning for Mean Monthly Rainfall Projections in Hawai'i." Journal of Artificial Intelligence for the Earth Systems (AIES), 2025 (submitted - addressing revisions).
3. O. Mudannayake*, A. Indika*, Janaka Alawatugoda, Upul Jayasinghe, and Gyu Myoung Lee. "On privacy-preserved machine learning using secure multi-party computing." Journal of Computers, Materials & Continua (CMC), 2025 (drafted - submitting soon).
4. (Best Paper Award) S. Jayasundara*, A. Indika*, and D. Herath. "Interpretable Student Performance Prediction Using Explainable Boosting Machine for Multi-Class Classification." 2022 2nd International Conference on Advanced Research in Computing (ICARC), pp. 391–396. DOI:
10.1109/ICARC54489.2022.97538675. A. Indika, P. Y. Washington, and A. Peruma*. "Performance Comparison of Binary Machine Learning Classifiers in Identifying Code Comment Types: An Exploratory Study." 2023 IEEE/ACM 2nd International Workshop on Natural Language-Based Software Engineering (NLBSE), pp. 20–23. DOI:
10.1109/NLBSE59153.2023.000126. A. Indika*, C. Lee, H. Wang, J. Lisoway, A. Peruma, and R. Kazman. "Exploring Accessibility Trends and Challenges in Mobile App Development: A Study of Stack Overflow Questions." 2025, 58th Hawaii International Conference on System Sciences (HICSS), Hawaii, USA. DOI:
10.48550/arXiv.2409.079457. N. Warusamana*, A. Indika*, E. Welikala*, and S. Deegalla. "Stock Market Prediction using SVM, LSTM, and Linear Regression." ESCaPe 2020 Project Symposium, pp. 21.
https://ESCaPE-2020-ProceedingsCurrently, I'm engaged in the following research projects:
I graduated as a Master of Science in Computer Science from Department of Information and Computer Sciences, University of Hawaii at Manoa, Hawaii (GPA: 3.91/4.00).
I graduated as a Bachelor of the Science of Engineering specialized in Computer Engineering with First Class Honours (GPA: 3.85/4.00).
I attended Maliyadeva College from Grade 12 to Grade 13 for my Advanced Level Education.
I attended Kegalu Vidyalaya from Grade 1 to Grade 11 during my Ordinary Level Education.
