Data Science professional with proven experience architecting automated data pipelines, optimizing robust relational databases, and deploying predictive machine learning models. Adept at using clean Python/SQL code to automate manual workflows and transform complex data structures into actionable insights. Skilled in end-to-end data analytics, validation protocols, and engineering interactive dashboards to drive data-driven decision-making and reporting.
Professional Experience
Electricité du Cambodge (EDC) — EDC-AFD-EU Project
Data Scientist Intern | March 2025 – Present
- Developed a meta-stacking ensemble machine learning model using temporal cross-validation to achieve high-accuracy predictive forecasting for electricity load requirements.
- Conducted advanced time-series exploratory data analysis (EDA) and engineered a feature selection framework, reducing data dimensionality from 151 down to 18 key indicators.
- Managed PostgreSQL schemas and optimized queries for the full-stack forecasting platform backend, ensuring seamless data consistency and real-time inference efficiency.
Ministry of Labour and Vocational Training (NCMW)
Data Analyst Intern | Aug 2025 – Oct 2025
- Built automated Python and SQL ETL pipelines to handle multi-source ingestion, replacing manual workflows and reducing data preparation time by ~70%.
- Implemented strict data validation protocols and root-cause analyses to locate errors and permanently eliminate system dataset inconsistencies.
- Designed 3 interactive Power BI dashboards mapping operational indicators, successfully utilized by team stakeholders for strategic daily decision-making.
Technical Projects Portfolio
Applied NLP Text Classification Engine (Email Verification Project)
- Architected an end-to-end processing pipeline to ingest, clean, and validate unstructured text datasets.
- Implemented tokenization and machine learning classifiers to execute text analytics, achieving a verified 98.5% classification tracking accuracy.
Anomalous Activity Detection & Financial Risk Modeling
- Analyzed massive, highly imbalanced transactional banking datasets to identify hidden anomalies, system risks, and inconsistent transaction footprints.
- Applied modern resampling frameworks (SMOTE) alongside Scikit-learn algorithms (Random Forest, Logistic Regression) to structure dependable risk models.
Volunteering
- Volunteer Trainer (Digital Literacy Program): Ministry of Education. Instructed 300+ students across 4 national high schools on core digital tool usage, online safety protocols, and operational data security practices (Mar 2025 – 2026).