Open to opportunities

Chhaya Athawale

@chhayaathawale

Data Scientist specializing in end-to-end ML and Generative AI, building production-ready automation across payments and supply chain.

India

Message

What I'm looking for

I’m looking for a role where I can build end-to-end ML and Generative AI (LLM/RAG) systems, deploy reliable APIs on Azure/AWS, and drive automation that reduces manual effort while improving model reliability.

I’m a Data Scientist with 5 years of experience building end-to-end ML and Generative AI solutions across Payments, Supply Chain, and E-Commerce, with hands-on production delivery. At VM Micro Solutions, I lead AI-driven automation that cut manual processing time by ~70% while engineering PySpark pipelines on Azure Databricks to deploy an XGBoost dispute risk model adopted by 500+ users.

I bring a strong focus on feature engineering, model evaluation, and scalable data pipelines—then push systems into real workflows through APIs and automation. I’ve integrated LLM capabilities into production using FastAPI and Azure services, including OCR-based invoice verification with human-in-the-loop workflows and an LLM-powered chatbot, and previously built live prediction APIs and serverless AWS pipelines for payment success/failure with measurable reliability improvements.

Experience

Work history, roles, and key accomplishments

Current

Data Scientist

Current

VM Micro Solutions Private Limited

Oct 2024 - Present (1 year 11 months)

Engineered Azure Databricks (PySpark) pipelines to build and deploy an XGBoost dispute risk prediction model for HP supply chain workflows. Designed and deployed an AI-powered travel management system using Python, FastAPI, PostgreSQL, and OCR/LLM features to reduce manual processing time by ~70%.

Dashboarding PySpark Azure Databricks XGBoost Feature Engineering Recall F1 Score Python fastAPI PostgreSQL Human In The Loop Duplicate Expense Detection Azure Blob Storage REST APIs

Python Data Scientist

Tech Mahindra

Apr 2021 - Sep 2024 (3 years 5 months)

Built and deployed ML models to predict real-time transaction success/failure for Lowe's payment gateway, integrating as a live prediction API to reroute high-risk payments. Architected an AWS serverless data pipeline and supported the full ML lifecycle, monitoring, and threshold tuning using A/B testing and confusion matrix analysis.