Open to opportunities

Yash Abhyankar

@yashabhyankar

Message

Data Analyst building KPI dashboards and predictive models that drive growth.

India

Message

What I'm looking for

I’m looking to help teams make faster, smarter decisions through automated ETL, KPI dashboards, and predictive models—owning the work end-to-end, collaborating closely, and delivering measurable business impact.

I’m a data analyst who turns complex business data into clear KPIs and reliable decision support. In my current role, I’ve engineered end-to-end Power BI dashboards by profiling 6,000+ records, writing complex DAX for 14 critical business KPIs, and optimizing ingestion by testing and automating vendor API workflows to cut manual data entry by 40%.

I also build scalable data foundations—Python-based ETL pipelines that normalize 50+ multi-sheet Excel/CSV workbooks with 100% data integrity, plus web scrapers that collect 10,000+ data points faster than manual extraction. Previously, I developed a customer churn model with Random Forest and Gradient Boosting (89.3% accuracy), used EKF to quantify customer behavior drift for better forecast precision, and contributed to a Condition-Based Monitoring + Data Lakehouse modernization effort, while my research focused on clustering and LLM prompting strategies.

Experience

Work history, roles, and key accomplishments

Current

Data Analyst

Current

CS Tech AI

Jun 2025 - Present (9 months)

Engineered 6 end-to-end Power BI dashboards tracking 14 critical KPIs across datasets of 6,000+ records. Built Python ETL pipelines for migrating and verifying 50+ Excel/CSV workbooks with 100% data integrity, and accelerated data collection by 3x using Playwright/Selenium scrapers.

Power BI DAX Python ETL PostgreSQL Selenium Data Migration

Executive (Data Analytics)

Solar Industries India Limited

Aug 2024 - Jun 2025 (10 months)

Built a customer churn model using 12 years of sales data, achieving 89.3% accuracy with Random Forest and Gradient Boosting, and added Isolation Forest anomaly detection for the top 10% of suspicious transactions. Used Extended Kalman Filter (EKF) to quantify customer behavior drift and improve sales trend forecast precision beyond baseline moving averages.

Churn Modeling Random Forest Gradient Boosting Isolation Forest Anomaly Detection Kalman Filter Forecasting Industry 4.0

Research Intern

University at Buffalo

Sep 2023 - Aug 2024 (11 months)

Researched DBSCAN/DENCLUE clustering methods and LLM architectures to improve prompt strategies for Small Language Models (SLMs). Focused on fine-tuning and prompting approaches to optimize model performance for downstream tasks.

DBSCAN DENCLUE Clustering LLM Prompts Optimization Machine Learning Neural Networks

Education

Degrees, certifications, and relevant coursework

University at Buffalo, SUNY

Master's in Data Science, Data Science

2022 - 2023

Completed a Master’s in Data Science at the University at Buffalo, State University of New York, studying data science concepts and applications through 2022–2023.