Shubham Pal

1. Leading Finance Transformation by deploying agentic AI systems (LangGraph, deep agents) to automate financial analysis, forecasting, and decision-making at scale. 2. Architected modular multi-agent workflows (EDA, modeling, evaluation) with dynamic orchestration, significantly improving speed and accuracy of financial insights. 3. Built and fine-tuned GenAI + RAG systems (PEFT/LoRA) for audit reporting and policy intelligence, reducing manual effort and standardizing outputs. 4. Enabled AI-powered analytics in Power BI, delivering real-time, query-driven insights for business leaders. Developed enterprise-scale cash flow & disbursement forecasting models used in executive planning and investor guidance. 5. Drove strategic decision-making on capital allocation and liquidity, contributing to multi-year stock growth and ~$75B increase in market cap through high-confidence forecasts. 6. Built Auto-Taxability ML system (AutoML + NLP/transformers) to classify tax applicability at scale. 7. Delivered ~$1M/month savings by eliminating overpayments and improving compliance. 8. Productionized ML systems using Python, PySpark, Azure Databricks, handling large-scale financial data pipelines. 9. Mentored teams and scaled adoption of agentic AI and AI-first engineering practices across org.

• Extracted, processed big datasets and designed deep learning models like GRU, LSTMs in python for time series forecasting of various rare events like hospitalization, FEP (First Episode of psychosis) for early-stage intervention program. • Used SMOTE, class weights and temporal sample weights for dealing with highly imbalanced class data problem. • Develop, design and deploy various machine learning algorithms like regression, negative binomial, decision tress, XGBoost for feature extraction and predictive analysis of clinical data. • Build text analysis pipeline using NLP, Word2Vec and BERT to cluster manually entered string records into meaningful groups. • Used SAS for basic statistical analysis of Medicaid data and other ad hoc requests. • Wrote PL/SQL queries to pull relevant data from MDW (Medicaid Data Warehouse) and monthly update tables and server migration.

• Extract, interpret and analyze data to identify key metrics and transform large raw data into relevant, actionable information using PL/SQL & SAS. • Gather data from MDW (Medicaid Data Warehouse) using Oracle SQL Developer according to data requests, Developed complex SQL queries using Window Functions , CTE and Optimized for performance using Views. • Analyzed network adequacy and impact of various healthcare service providers. • Collect, cleanse and provide modeling and analyses of structured and unstructured data for Office of Mental Health. • Wrangling ,cleaning and merging big datasets using Pyspark and Python Dask for creating a feeder database for tableau dashboards. • Create visually impactful dashboards in Tableau for data reporting using roll-up tables and publish them on server, also developed similar visualizations using Python Matplotlib and Seaborn. • Create reports to share findings and recommendations with the internal team and other stakeholders.

-Engineered real-time financial reports & dashboards in Tableau for NY’s residential and commercial projects. -Spearheaded multiple analyses, devising predictive models & algorithms on a wide range of key metrics using SAS ,SQL and Python. -Presented insights & strategy recommendations to project managers, team leads, & executive management. -Manipulated big data using PySpark and Dask to provide queries, statistical summaries, and data visualizations. -Forecasted consumption / engagement & identified consumer preferences to augment social media outreach. -Built Predictive models for customer segmentation and power consumption cost analysis to understand customer energy consumption behaviour using Sklearn, Python and R.

* Developed a machine learning system that predicted purchase probability at a particular offer based on customer’s real time location data and past purchase behaviour. • Performed k-Means clustering in order to understand customer itemized bought products and segment the customers based on the customer products for animal medicine and vaccines behaviour information for customized product offering, customized and priority service, to improve existing profitable relationships and to avoid customer churn, etc. using Python. • Used PySpark Machine learning library to build and evaluate different models. • Generated, wrote and ran SQL script to implement the database changes including table update, addition or update of indexes, creation of views and store procedures. • Developed advanced SQL queries with multi table joins, groups, functions, sub queries, set operations, & stored procedures for Data Analysis in SQL server. • Programmed ETL functions between Oracle and Amazon Redshift. Managed big data files in AWS S3 using boto3 client. • Developed and maintained stored procedures and complex packages extensively using PL/SQL and shell programs.