Looking for a job

Romith Reddy

@romithreddy

Message

I test OCR models and validate data quality through dashboards, preprocessing, and accuracy analysis.

Canada

Message

What I'm looking for

I'm looking for an opportunity where I can validate data and model outputs, improve data quality, build clear reporting dashboards, and collaborate with cross-functional teams on practical analytics and automation.

At SumPositive LLC, I test OCR models on simulated healthcare claim forms and compare extracted results against structured ground-truth data for fields including PHIN, DOB, and tariff codes.

I've evaluated Google Vision API, Tesseract OCR, EasyOCR, PaddleOCR, TrOCR, Meta-LLaMA, and Amazon Textract, identifying extraction issues and supporting preprocessing improvements. I also contribute to an OCR processing pipeline integrated with App Center and collaborate with Claims, Data, and Engineering teams on validation rules.

I'm building Power BI dashboards for model-performance reporting and have improved a disease-detection model's accuracy from approximately 25% to 54% through feature engineering and data transformation. My work combines Python, SQL, Excel, data validation, dashboard testing, and practical analysis to improve reliable decision-making.

Experience

Work history, roles, and key accomplishments

Current

QA / Data Analyst Intern

Current

SumPositive LLC

Nov 2025 - Present (10 months)

Evaluated multiple OCR models on simulated healthcare claim forms, designed ground truth datasets, and performed detailed validation of model outputs. Built Power BI dashboards to visualize model performance and collaborated with cross-functional teams to improve data extraction accuracy.

GoogleVision API Tesseract OCR EasyOCR Paddle OCR TROCR Meta’S LLaMA Amazon Textract Power BI Data Validation Model Testing data preprocessing OCR Pipelines Python SQL Excel

Data Analysis & Model Validation Project

Mohawk College

Sep 2024 - Jun 2025 (9 months)

Developed interactive Power BI dashboards for early disease detection, performed data preprocessing, and built machine learning models. Improved model accuracy from ~25% to ~54% through feature engineering and validated predictions using performance metrics.

Power BI data preprocessing Machine Learning Logistic Regression Feature Engineering ANOVA RFECV Neo4j Streamlit Python

PYROGENESIS Market Research and Strategic Planning Project

Niagara College

May 2024 - Aug 2024 (3 months)

Supported financial report development and market analysis with clean data sets, and created portfolio summaries and performance reports for business planning presentations.

Financial Reporting Market Analysis Data Cleaning Business Planning

Project 2 - Reduce Obstacles for the Adoption of Advanced Technologies

Mohawk College

Created Excel-based reports and visualizations for market and performance data, and collaborated across teams to translate data into actionable recommendations.

Excel Data Visualization Statistical Analysis Collaboration