# Hello World, I am

Gayatri Patil

AI & Data Scientist

View My Work
Gayatri Patil

Professional Experience

Kushmanda.ai

Product Data Science Intern
GoLang Apache Druid React-Native
  • Designed GoLang APIs for data onboarding, enabling ingestion of 3 structured/semi-structured sources.
  • Integrated Apache Druid for real-time OLAP queries, improving latency by 25%.
  • Built cross-source joins and React-Native visualization features for enhanced reporting.

Michelin

Developer Analyst
Azure Data Lake ETL/ELT SQL
  • Deployed 20 ETL/ELT pipelines facilitating data flow across 4 cross-functional teams.
  • Managed Workday HCM ingestion into Azure Data Lake with 100% governance compliance.
  • Automated 20+ reports, increasing adoption by 19% and optimized SQL performance.

LTI - Larsen & Toubro Infotech

Software Engineer
Java REST APIs SonarLint
  • Developed KYC automation software, reducing processing time from 7 days to 4 hours.
  • Raised data accessibility by 25% by developing backend RESTful APIs.
  • Reduced production bugs by 10% through SonarLint initiatives and code refactoring.

Featured Projects

HelioSync

Federated Learning MLOps
Docker Flask TensorFlow PyTorch
  • Architected a modular FL platform with Dockerized services for client orchestration and secure aggregation (FedAvg).
  • Implemented orchestration layer with CI/CD & visualization ensuring fault-tolerant training across clients.

Beyond the Tap

Water Quality Monitoring
LightGBM UMAP K-means
  • Achieved >90% accuracy in classifying water quality for contamination detection using ML techniques.
  • Applied feature engineering and dimensionality reduction for interpretable, reliable solutions.

Adaptive AI Agent

Browser Automation Architecture
React FastAPI LLM Playwright
  • Designed a Browser Agent (ABA) to reduce latency/cost and overcome brittle action grounding in LLM agents.
  • Integrated perception, memory, and sharing modules to turn successful interactions into reusable capabilities.
  • Developed a multi-agent pipeline: User Query -> ProductSearch -> PriceComparison -> Recommendation.

Pitch-Perfect

Cricket Shot Analysis
TensorFlow FastAPI React CV
  • Built a full-stack app for real-time cricket shot classification and form quality assessment using Deep Learning.
  • Implemented a multi-task model to classify shots (drive, pull) and assess form quality with confidence scoring.
  • Enabled real-time analysis supporting both live camera feeds and video file uploads.

GitSight

GitHub BigData Analytics
PySpark Snowflake Azure Power BI
  • Engineered a distributed pipeline to process 54GB+ GitHub metadata, analyzing developer influence and trends.
  • Structured raw activity data into Snowflake, creating Power BI reports to aid recruiters and product teams.

Weather Analytics

Data Pipeline Orchestration
Airflow Snowflake DBT Tableau
  • Built a scalable ETL pipeline to ingest and process weather data for agriculture and urban planning insights.
  • Orchestrated data flows using Apache Airflow and DBT, visualizing key metrics in Tableau.

Technical Arsenal

Languages

Python SQL Java R GoLang

AI/ML

GenAI RAG Computer Vision PyTorch TensorFlow

Tools & Cloud

AWS Azure Docker Airflow Snowflake

Education

San José State University

M.S. Applied Data Science

Aug 2024 – May 2026 | California, USA

Savitribai Phule Pune University

B.E. Computer Engineering

July 2015 – June 2019 | Pune, India

Awards & Recognitions

Go-Getter Award

Awarded by Michelin, 2023