Yash Raythatha

About Yash Raythatha

Ever since I immersed myself in the realm of data, it has been an exhilarating journey of discovery. In the modern industrial landscape, data is considered the new oil. Just as unrefined oil serves no purpose, unrefined data lacks utility. I derive joy from unraveling complex data sets and transforming them into meaningful, actionable insights.Over the course of my 4+ year career, I have been fortunate to collaborate with a diverse range of organizations, both large and small. I have applied my skills to address a wide spectrum of challenges. During this time, I have honed my proficiency in Python, SQL, data extraction, data loading, data cleaning, data visualization, and ETL processes. Additionally, I have a keen interest in cloud computing and hold certifications as an Azure Administrator (Az-104 and Az-900) and Oracle SQL Database (1Z0-071). Lastly, I also hold the Power BI Data Analyst Associate (PL-300) Certification.Currently, I am a student at the University of New Haven, pursuing a master\'s degree in data science to refine my skills and align them with industry standards.Some additional facts about me include-My passion for continuous learning led me to leave my full-time job and pursue a Master\'s in the USA. I am driven and refuse to settle-I prioritize making quality decisions-I excel in communication and enjoy collaborating with team members to achieve common goals.Lastly, beyond the realm of data, I am an avid enthusiast with diverse interests. Whether passionately watching cricket, binge-watching movies or series on Netflix, exploring new places through travel, immersing myself in music, or engaging in meaningful conversations, I find joy in various pursuits.If you share similar passions or have exciting new ideas, I\'d love to connect!kindly visit my portfolio website : https://yashraythatha.com/

Experience

1. Ai Data Analyst

   Gratiture Solutions

   Aug 2025 — Present

   AI Hackathon Code Evaluation SystemBuilt a multi-agent AI evaluation engine using Python, CrewAI, LangChain, and OpenAI to automate code assessment across architecture, ML logic, UI/UX, security, and performance. Designed a hybrid LLM and heuristic scoring pipeline that generated structured technical feedback with code references, eliminating about 85 percent of manual scoring effort and reducing evaluation costs by roughly per month.Synthetic Data Generation PlatformDeveloped a privacy-first synthetic data engine using SDV, FAISS embeddings, and Presidio for PII detection with k-anonymity validation. Ensured PII-free compliance, reduced ML prototyping cycles by approximately 70 percent, and saved around per year by removing dependence on licensed datasets and manual privacy workflows.Digital Course Data PlatformBuilt a full-stack course catalog and enrollment platform using React, TypeScript, Node.js, and PostgreSQL supporting over 300 courses and labs. Optimized schemas and APIs to improve query performance by about 45 percent, deployed on Azure with CI/CD pipelines, and created an admin console with bulk Excel import and export that reduced manual administration by 60 percent and saved about per month.Adaptive Question Selection SystemDesigned an adaptive learning platform using a hybrid architecture to personalize exam questions in real time. Built a high-performance rule engine with under 10 ms decision latency, integrated GPT-based reasoning for intelligent topic prioritization, and deployed the system as a FastAPI microservice with MySQL and real-time frontend integration.Usage-Based SaaS Pricing EngineDeveloped a usage-based pricing and forecasting engine using TypeScript, Node.js, and PostgreSQL. Automated pricing rules and scenario modeling, reducing manual pricing analysis by 80 percent and saving approximately per quarter.

Education

• University of New Haven

  Master's degree, Data Science

• Rajiv Gandhi College of Engineering, Research and Technology

  Bachelor of Engineering - BE, Computer Technology/Computer Systems Technology