Ra Lan Đỗ Tú Trinh

Key Skills

Software

Data Annotation Tech

Internal/Proprietary Tooling

Top Subject Matter

No subject matter listed

Top Data Types

Audio

Document

Top Label Types

Classification

Transcription

Data Collection

Freelancer Overview

I am a Data Science student with hands-on experience in building multi-modal AI systems that integrate satellite imagery and spatiotemporal sensor data, particularly for environmental resilience and hazard forecasting. My work involves developing and optimizing data pipelines, normalizing large-scale atmospheric telemetry, and annotating complex geospatial datasets for use in deep learning models such as LSTM and GRU. I have a strong background in data labeling and feature engineering, utilizing tools like Python (Scikit-learn, Pandas, NumPy), PyTorch, ArcGIS, and QGIS. My projects emphasize explainable AI, using SHAP and LIME to ensure model interpretability and consistency with real-world phenomena. I have also led research and participated in high-impact competitions, where I designed workflows for processing and annotating multi-parameter environmental data, demonstrating my ability to manage and quality-check diverse training datasets for AI applications.

Labeling Experience

Research Assistant

Internal Proprietary ToolingDocumentData Collection

- Conducted comprehensive fact-checking and background research on prominent academic figures associated with USC. - Verified and updated the status and biographical data of distinguished professors to ensure the integrity of academic databases. - Cross-referenced multiple scholarly sources and institutional records to maintain high data accuracy. - Managed and organized sensitive information regarding academic legacies and professional histories.

- Conducted comprehensive fact-checking and background research on prominent academic figures associated with USC. - Verified and updated the status and biographical data of distinguished professors to ensure the integrity of academic databases. - Cross-referenced multiple scholarly sources and institutional records to maintain high data accuracy. - Managed and organized sensitive information regarding academic legacies and professional histories.

2025

Research Assistant Labelling

Data Annotation TechAudioClassificationTranscription

- Transcribed high-quality audio datasets for linguistic research and AI training. - Filtered and screened audio files to identify specific speaker profiles (e.g., excluding MC voices) based on project requirements. - Classified speech by regional dialects and identified environmental background noise (ambient noise, interference) to ensure data accuracy. - Maintained high precision in converting spoken language into written text while following strict formatting guidelines.

- Transcribed high-quality audio datasets for linguistic research and AI training. - Filtered and screened audio files to identify specific speaker profiles (e.g., excluding MC voices) based on project requirements. - Classified speech by regional dialects and identified environmental background noise (ambient noise, interference) to ensure data accuracy. - Maintained high precision in converting spoken language into written text while following strict formatting guidelines.

2025 - 2025

Education

Work History