Data Engineer | Big Data & BI | MLOps | Full-Stack Developer

Joined as Full-Stack Developer & Data Engineer to develop and maintain business applications and internal software solutions.

• Designed FastAPI backend services and PostgreSQL database structures for scalable internal tools.
• Contributed to digital transformation and ERP initiatives in collaboration with technical and business teams.
• Technical environment: FastAPI, PostgreSQL, Python, Docker, version control, Linux.

Designed and deployed an end-to-end demand forecasting and supply planning system in production, reducing planning effort and improving order accuracy.

• Developed a LightGBM forecasting model across 52 SKUs with 12-week horizon, achieving 47% median WAPE on 2025 test data, in hybrid architecture combining confirmed orders (29 SKUs, 16.7M units, 100% accuracy) with ML for uncovered SKUs.
• Implemented complete MLOps pipeline: PSI drift detection, 4 automated governance rules, champion/challenger strategy, MLflow model registry, and automated alerts (MODEL_DEGRADED, DATA_DRIFT, OVERDUE_ORDER).
• Automated pipeline with APScheduler to run weekly without human intervention.
• Built KPIs and dashboards in React + charting library (forecast views, realized WAPE tracking, PSI charts, governance logs).
• Developed AI assistant (open-source LLM, ReAct tool calling, 6 database-connected tools) answering natural language questions on inventory, forecasts, and alerts.
• Technical environment: Python, LightGBM, FastAPI, PostgreSQL, MLflow, React, containerization, task scheduling, open-source LLM.

Joined data team as Big Data Engineer Intern to automate data ingestion and process monitoring.

• Automated ingestion and monitoring of spreadsheet files with Python.
• Built robust ETL to clean, merge, and load data into PostgreSQL.
• Developed business intelligence dashboards and optimized ETL/pipeline workflows.
• Technical environment: Python, PostgreSQL, business intelligence tools, ETL/SSIS.

Worked as Python Developer to build scraping pipelines and perform sentiment analysis.

• Built social media scraping pipelines and performed sentiment analysis in local dialect.
• Automated data collection, preprocessing, and report generation.
• Technical environment: Python, browser automation, NLP, sentiment analysis.

Developed web and mobile monitoring applications using modern full-stack technologies.

• Developed web and mobile applications for quality monitoring using React.js, React Native, Flask, PostgreSQL and ML.
• Technical environment: React.js, React Native, Flask, PostgreSQL, machine learning.