📍 Chicago, IL

Hello, I'm Gowtham — Data Scientist

4+ years building predictive analytics, statistical models, and scalable machine learning & AI systems across financial services, healthcare, and retail. I turn large, messy datasets — and increasingly, LLM-powered agents — into clear, production-ready decisions.

View My Work Download Resume

gowthamkpiratla@gmail.com LinkedIn

Featured Work

Recent Projects

Two end-to-end systems — data engineering and multi-agent AI — built and documented from scratch.

Retail Sales ELT Pipeline

Raw Data → Analytics-Ready Star Schema

An end-to-end ELT pipeline that turns a messy raw sales export into a query-ready dimensional model — automated cleaning of mixed date formats, duplicate orders, missing prices, and hidden returns, partitioned Parquet storage, and an Airflow-orchestrated daily run feeding a DuckDB/BigQuery-ready star schema for revenue, product, and customer analytics.

PythonPandasSQL DuckDBParquetAirflow

Code →

Enterprise AI Decision Intelligence Platform

Multi-Agent AI System for Automated Business Investigation

A multi-agent AI platform — 8 agents orchestrated with LangGraph — that automates root-cause business investigations: validating data quality, running statistical significance tests, forecasting expected performance, decomposing root cause across business segments, and generating an executive-ready summary. Turns a multi-day manual analyst workflow into a sub-minute automated one, with a FastAPI backend, Streamlit UI, and PPTX/PDF export.

PythonFastAPIStreamlit LangGraphOpenAI APIScikit-learn statsmodelsDocker

Code →

Career

Experience

4+ years shipping analytics, ML, and AI in production.

Jan 2025 – Present

Data Scientist

Wells Fargo · USA

Developed and deployed predictive ML models (Python, Scikit-learn, XGBoost) improving fraud detection accuracy by 25% across enterprise risk operations.
Built production-ready NLP and LLM-powered pipelines with LangChain and OpenAI APIs, boosting insight-extraction efficiency by 40% on identity and risk workflows.
Automated executive Power BI dashboards and KPI scorecards with DAX and SQL, cutting reporting turnaround by 20%.
Partnered with data engineering to optimize ETL workflows on AWS and Azure, improving pipeline reliability by 35% across 5+ production systems.

Sep 2020 – Aug 2023

Data Scientist

Hexaware Technologies · India

Analyzed 10M+ record datasets with Python and SQL, improving predictive model performance by 18% through statistical inference and feature engineering.
Built classification, regression, clustering, and recommendation models (Scikit-learn, TensorFlow, PyTorch), improving analytical accuracy by 22%.
Developed Tableau dashboards translating ML-driven insights into business decisions across multiple units.
Collaborated with data engineering to deploy models to production and maintain ongoing monitoring and optimization.

Toolbox

Technical Skills

The tools I reach for every day.

Programming & Querying

PythonSQLRScala

ML & Data Science

Predictive AnalyticsStatistical ModelingNLPForecastingClassificationClusteringRecommendation SystemsA/B Testing

AI & Generative AI

LLMsRAGPrompt EngineeringLangChainHugging FaceOpenAI APIs

Frameworks & Libraries

Scikit-learnTensorFlowPyTorchXGBoostPandasNumPyKeras

Cloud & Big Data

AWSAzureDatabricksSparkSnowflakeHadoopKafka

MLOps & Data Engineering

MLflowAirflowDockerKubernetesETL PipelinesCI/CD

Visualization & BI

Power BITableauMatplotlibSeabornExcel

Databases

PostgreSQLMySQLMongoDBRedis

Master's in Computer Science

DePaul University · Chicago, IL · Sep 2023 – Nov 2025

About Me

A bit about my work

I'm a Data Scientist based in Chicago, IL, with 4+ years of experience designing predictive analytics, statistical models, and scalable ML solutions across financial services, healthcare, and retail. I like the messy middle of data work — understanding the business problem, shaping the data, and turning the result into something an executive can actually act on.

My day-to-day spans Python, SQL, and the modern ML stack — Scikit-learn, TensorFlow, XGBoost — along with distributed processing on Spark and cloud-native workflows on AWS and Azure. I'm especially energized by generative AI: building RAG systems and multi-agent workflows with LangChain, OpenAI APIs, and vector databases that move organizations from static dashboards to real decision intelligence.

I'm currently pursuing my Master's in Computer Science at DePaul University while working as a Data Scientist at Wells Fargo, building fraud analytics, NLP workflows, and executive reporting at scale.