Prabash

Engineered end-to-end data pipelines in Python to ingest, validate, and preprocess 5,000+ camera system images, implementing automated data quality checks that reduced manual preparation time by 30% and accelerated downstream ML evaluation cycles. Executed data extraction and wrangling across relational databases using advanced SQL and EDA to extract 10+ operational KPIs (defect rates, processing efficiency, system latency), delivering structured datasets and actionable recommendations that informed product strategy. Developed statistical analysis framework using t-tests and chi-square tests to validate performance improvements across camera system configurations, enabling data-driven decision-making for AI/ML deployment prioritization.

Data Analyst Intern at Samsung (2025-09 – 2026-02)

• Engineered end-to-end data pipelines in Python to ingest, validate, and preprocess 5,000+ camera system images, implementing automated data quality checks that reduced manual preparation time by 30% and accelerated downstream ML evaluation cycles.
• Executed data extraction and wrangling across relational databases using advanced SQL and EDA to extract 10+ operational KPIs (defect rates, processing efficiency, system latency), delivering structured datasets and actionable recommendations that informed product strategy.
• Developed statistical analysis framework using t-tests and chi-square tests to validate performance improvements across camera system configurations, enabling data-driven decision-making for AI/ML deployment prioritization.
• Automated REST API data extraction workflows processing 1,000+ records per pipeline run, implementing error-handling and retry mechanisms to ensure consistent, audit-ready data integrity across 5+ business domains.
• Designed interactive dashboards using Plotly and Power BI that translated complex camera system test metrics into clear business narratives, tracking quality KPIs and root-cause analysis across 8+ edge-case scenarios.
• Collaborated with cross-functional teams to identify data quality gaps and systematized reporting processes, reducing manual data consolidation time by 40% and enabling real-time stakeholder visibility into system performance metrics.

Data Engineer Intern at Information Retrieval Lab IIT(BHU) (2025-05 – 2025-07)

• Designed and deployed end-to-end ETL pipelines for multilingual text preprocessing, executing automated cleaning, schema consistency checks, and feature enrichment to accelerate data readiness for predictive modeling and model training.
• Built advanced evaluation dashboards using Python to track transliteration quality metrics (WER/CER); executed slice-level error analysis by token length, OOV rate, and script-mixing to pinpoint algorithmic failure modes.
• Applied chi-square tests and t-tests to validate statistical significance of WER/CER metric improvements across model runs.
• Built business intelligence and self-service reporting frameworks integrating metric trends and statistical significance testing (t-tests, chi-square); translated complex data insights into compelling data stories for technical and non-technical stakeholders.

M.Tech Integrated in CSE — Data Science – VIT, Vellore (2022-07 – 2026-06)