Looking for a job

Shivam Mali

@shivammali

I build predictive machine learning systems for healthcare, banking, churn, forecasting, and transaction classification.

India

Message

At EXL Services, I build healthcare revenue and claims-cost prediction models for CVS Health Aetna, improving accuracy by approximately 5–6% over actuarial GLMs. I productionize scalable workflows with Python, SQL, PySpark, and Airflow, supporting pricing, reserve planning, and multi-million-dollar financial outcomes.

Previously, I delivered banking models at 3LOQ Labs for HDFC Bank, including an XGBoost and RandomForest model with 88% recall and a transaction classification engine that improved accuracy from 83% to 99%. I've also built churn, reward optimization, market-share, and revenue forecasting solutions, and developed a Google ADK RAG/NL2SQL agent to make model predictions easier for actuaries to understand.

Experience

Work history, roles, and key accomplishments

Current

Data Scientist

Current

EXL Services

Sep 2024 - Present (2 years)

Developed and deployed machine learning models to predict healthcare revenue and claim costs for CVS Health Aetna, improving accuracy by 5-6% over traditional actuarial models. Automated pipelines with Airflow, PySpark, SQL, and Python.

Python PySpark SQL Airflow LightGBM XGBoost Gradient Boosting SHAP Google ADK RAG NL2SQL

Data Scientist

3LOQ Labs PVT LTD

Mar 2023 - Sep 2024 (1 year 6 months)

Developed predictive models for HDFC Bank to identify customers likely to add funds within three months, achieving 88% recall. Also built a credit engine to classify transactions with accuracy improved from 83% to 99%.

Python XGBoost Random Forest Airflow Bag Of Words Cosine Similarity

Data Scientist

Innominds Software PVT LTD

Feb 2022 - Dec 2022 (10 months)

Developed an application for churn prediction and total reward optimization, integrating multiple customer datasets. Extracted data from AWS S3, performed EDA, and built predictive models using XGBoost, RandomForest, and H2O.ai.

Python XGBoost Random Forest H2O Airflow S3

Engineer

Aarvi Encon

Jan 2020 - Dec 2020 (11 months)

Developed machine learning models to predict market share and Net Revenue Realization growth. Performed data preprocessing and analysis using Python, and implemented predictive models including Linear Regression, XGBoost, and SARIMA.

Python H2O XGBoost Linear Regression SARIMA