Amalsha Fernando

Built a 3,000+ sample Sinhala speech dataset and fine-tuned the F5-TTS Base model to develop a Sinhala Text-to-Speech system. Developed the end-to-end TTS pipeline, including data preprocessing, audio normalization, model training, and evaluation.

Developed multi-agent communication systems integrating voice, email, and messaging APIs to automate client interactions. Built an AI-powered SQL Assistant and designed large-scale SEO optimization systems with traffic simulation capabilities.

Designed and implemented Retrieval-Augmented Generation (RAG) systems using vector databases for efficient information retrieval. Developed and fine-tuned deep learning models through feature engineering and exploratory data analysis.

Pursuing a Master of Artificial Intelligence, expected to complete in 2026.

Currently pursuing the Deep Learning Specialization offered by Stanford University.

Completed a BSc (Hons) in Information Technology with a specialization in Information Technology.

Completed the Machine Learning Specialization offered by Stanford University.

Earned the Azure Fundamentals Certification from Microsoft.

Completed AI ML Engineer Stage 1 and Stage 2 certifications from SLIIT.