SAINATH KONDRI
Kent,OH
Summary
Results-driven Machine Learning Engineer with 3+ years of experience in developing data-intensive solutions for financial risk and digital advertising. Built feature pipelines and classification models while processing large datasets using SQL and PySpark. Delivered production-ready ML workflows through APIs, Docker, and CI/CD, effectively translating business challenges into measurable machine-learning systems.
Overview
4
4
years of professional experience
Work History
Machine Learning Engineer Intern
Charles Schwab
08.2025 - 07.2026
- Developed a transaction-risk scoring model using XGBoost and Scikit-learn on multi-million-record historical transaction and customer-behavior datasets to identify potentially high-risk activity for downstream fraud review.
- Built PySpark and SQL feature pipelines generating behavioral, velocity and historical aggregation features—including transaction frequency, amount deviation and recent activity patterns—for model training and validation.
- Improved fraud recall by ~8% at comparable false-positive rate through feature engineering, class-imbalance handling, hyperparameter tuning and decision-threshold optimization.
- Performed false-positive/false-negative analysis and SHAP-based interpretation to identify key risk drivers, diagnose model errors and support discussions with risk and analytics stakeholders.
- Tracked experiments, model configurations and evaluation artifacts with MLflow, improving reproducibility across model iterations and supporting model-version comparison and monitoring workflows.
- Collaborated with data scientists, engineers and risk stakeholders to translate fraud-detection requirements into model features, evaluation criteria and operational alert thresholds.
Machine Learning Engineer
HCL Tech - India
04.2022 - 07.2024
- Developed audience-targeting and engagement-prediction solutions using Python, SQL, Scikit-learn and PySpark across tens of millions of historical impression, click, conversion and customer-interaction records.
- Built reusable PySpark ETL pipelines to clean, join and aggregate campaign and behavioral event data into customer- and campaign-level training datasets.
- Built gradient-boosted and engagement/promotion models and improved validation ROC-AUC from ~0.71 to ~0.76 through behavioral feature engineering, cross-validation and hyperparameter optimization.
- Developed behavioral and engagement features from historical user interactions to support audience segmentation and recommendation/personalization workflows.
- Evaluated targeting strategies through A/B testing using CTR, conversion rate and engagement metrics, with successful variants delivering approximately 7.5% relative improvement in conversion rate.
- Built SQL analytical workflows and Power BI dashboards to monitor campaign performance, conversion trends and customer engagement, enabling stakeholders to investigate underperforming campaigns and targeting segments.
Education
Master of Science - Computer Science
Kent State University
Kent, OH05-2026
Skills
- Machine learning
- Programming: Python, SQL
- Data: Pandas, NumPy, PySpark, ETL Pipelines
- MLOps tools
- A/B testing
- Databases: PostgreSQL, MySQL, Snowflake
- Backend/Cloud: FastAPI, REST APIs, AWS
- Visualization: Power BI, Streamlit
Timeline
Machine Learning Engineer Intern
Charles Schwab
08.2025 - 07.2026
Machine Learning Engineer
HCL Tech - India
04.2022 - 07.2024
Master of Science - Computer Science
Kent State University