SUYASH.

Senior Data Scientist | Architecting AI-Powered Experimentation Systems

• Architected an end-to-end Causal Experimentation Engine: Replaced manual workflows with a high-velocity pipeline scaling from 3 to 8+ experiments/week.
• The Pipeline: Engineered a robust system: Test Configs (Experiment ID) → Automated Exposure Tables → 31+ Metric Aggregation → CUPED Variance Reduction → Bayesian Inference Layer.
• Statistical Impact: Reduced standard error by 35%, compressing 28-day test windows into 12 days.
• GenAI Integration: Deployed a GenAI Analyst Agent (SQL+RAG) and an Auto-Summary Agent, reducing post-experiment analysis from days to hours while cutting hallucinations by 60%.
• Business Value: Validated $150M+ in annual strategic decisions across Pharmacy, Digital, and Private Brands.

Principal Engineer & Product Builder

Founded an applied research lab to validate emerging Agentic and RAG architectures before enterprise deployment.

• QuizBeef (700+ Users): Architected an AI quiz app utilizing LLM semantic parsing and FAISS vector retrieval for document comprehension; insights informed Albertsons' GenAI Copilot patterns.
• mana-health: Developed a multi-agent health platform orchestrating specialized agents (physio, nutritionist) to process biometric and workout data.
• Aligned: Built a dating app integrating astrology compatibility after market analysis revealed an underserved segment; demonstrates product-market fit analysis and consumer behavior insights.

Senior Data Scientist

• $114M in annual loan volume by leveraging XGBoost analysis to prove employment verification added no predictive value, then executing an A/B test on low-risk segments that increased conversions by 300+ customers monthly.
• $1.8M in loan approvals by identifying and resolving critical iOS bugs through funnel analysis on the Automated Loan Approval platform.
• Improved customer segmentation model accuracy by 12% by integrating GCP TensorFlow and collaborating with ML engineering team.
• Reduced infrastructure costs by 12+% by migrating legacy SAS workflows to modular SQL queries.

Data Science Engineering Intern

• Informed 3+ marketing campaigns by analyzing user clustering patterns to uncover distinct customer journey types.
• Reduced authentication false positives by 13% by building ETL pipelines and deploying an anomaly detection model on AWS SageMaker.
• Improved forecasting accuracy by 8.5% by designing a hybrid time-series model (Prophet + LSTM) for sales prediction.

Assistant Manager, Data Science

• $11.2M marketing revenue increase and improved CTR by 28% by orchestrating email personalization using logistic regression.
• Enhanced customer lifetime value by 5% by developing Linear Regression models to optimize subscription pricing.
• Reduced Priority 2 production issues by 60+% by developing Tableau dashboards with time series forecasting alerts.

The University of Texas at Dallas

MS, Business Analytics • Jun 2022

Dean's Scholar (Top 20%); President, Data Science Club.

Pune University

BS, Computer Science and Engineering • Jun 2016

Programming & Data

Python, SQL, PySpark, BigQuery, Airflow, Pandas, NumPy, Scikit-learn, TensorFlow, LangChain

Machine Learning & Product

A/B Testing, Bayesian Methods, Causal Inference, Uplift Modeling, Prompt Engineering

Infrastructure

GCP (BigQuery, Vertex AI), AWS (S3, SageMaker), Azure, Docker, Streamlit

Certifications

Google Analytics