Sinhala Text Translation & Annotation
Annotated and translated Sinhala texts into English for AI training datasets. Ensured semantic accuracy and consistency to enhance multilingual model performance.
Hire this AI Trainer
Sign in or create an account to invite AI Trainers to your job.
No subject matter listed
I’m an AI researcher and computer science student with practical experience in data annotation, model evaluation, and AI training workflows. Through my Oracle Cloud Infrastructure (OCI) Data Science and Generative AI Professional certifications, I gained hands-on experience in preparing and labeling datasets using tools like Label Studio, CVAT, Roboflow, and Doccano. My work covered text classification, image labeling, entity tagging, and prompt–response evaluation for fine-tuning and RLHF tasks. I’ve applied these skills in multiple projects, including a Smart India Hackathon solution that used computer vision to identify cows and assist farmers, and another project that detected Sinhala text in images and translated it into English using OCR and language models. These experiences strengthened my understanding of how high-quality, consistent annotation directly impacts model performance. I’m passionate about building structured datasets that bridge human understanding with machine learning — ensuring fairness, precision, and real-world usability in AI systems.
Annotated and translated Sinhala texts into English for AI training datasets. Ensured semantic accuracy and consistency to enhance multilingual model performance.
Labeled images of cows to train an AI model for instant identification, helping farmers track and manage livestock. Focused on accurate annotation and quality checks to ensure model reliability.
Developed a web-based Streamlit application to facilitate Reinforcement Learning from Human Feedback (RLHF) for LLM code generation. The tool was used to collect human evaluations and ratings on Python code generated by multiple LLM providers, with the goal of improving the models over time. This project was completed for a college assessment.
Annotated and classified video datasets for Oracle AI training projects, including action recognition and quality evaluation. Ensured high-quality labeling to improve model accuracy and gained hands-on experience with video-based AI workflows.
High School, High School
Integrated MCA, Computer Science
Founder
Founder