Professional Experience

AI Engineering Intern @ Infosys Springboard

Tech Stack: LangGraph, Google Gemini, FastAPI, Playwright, React

• Architected an end-to-end AI-powered automated testing platform using LangGraph, Google Gemini, FastAPI, and Playwright; managed full AI lifecycle from use-case scoping to production deployment covering 10+ real-world scenarios.
• Engineered an LLM orchestration pipeline translating natural language test instructions into precise DOM assertions, reducing manual QA scripting time by 40%.
• Designed a low-code React frontend with real-time execution feedback (pass/fail metrics, step logs, screenshot captures) for non-technical stakeholders.
• Benchmarked Gemini model variants (2.0-flash, 2.5-flash) against latency and accuracy KPIs, documenting findings to optimize AI performance.

Applied Machine Learning Intern @ IBM SkillsBuild – Edunet Foundation (AICTE)

Tech Stack: Next.js, FastAPI, XGBoost, Docker, GitHub Actions

• Architected HydrAI, a full-stack AI water demand prediction platform (Next.js, FastAPI, XGBoost); achieved R² > 0.80 on strict chronological validation.
• Built a fault-tolerant Gemini integration with a multi-model fallback waterfall and exponential back-off, maintaining near-100% LLM uptime for reliable AI responses.
• Containerized the application with Docker Compose and built CI/CD pipelines via GitHub Actions, integrating in-memory hash-keyed caching to reduce API costs.
• Evaluated ML pipelines for customer churn prediction (Logistic Regression vs. Random Forest) and presented findings as an executive-ready analytical report.

Power BI Intern @ Microsoft Elevate

Tech Stack: Power BI, Data Analytics, ETL workflows

GitHub: IPL-Dashboard

• Built interactive dashboards using Power BI for comprehensive business intelligence use cases.
• Performed data cleaning, transformation, and modeling on real-world datasets to ensure analytics readiness.
• Applied advanced visualization techniques to communicate hidden insights effectively to stakeholders.
• Strengthened understanding of holistic analytics workflows and data-driven decision systems.

Data Analyst Intern @ VOIS

Tech Stack: SQL (CTEs, Window Functions), Python, Pandas

• Performed advanced SQL analysis using CTEs, window functions, and multi-table joins on large telecom datasets to derive actionable insights.
• Conducted Pandas-based EDA to consistently identify and measure key customer churn drivers.
• Automated ETL data-cleaning workflows, successfully reducing manual reporting effort by 30%.
• Translated complex analytical findings into concise executive presentations and business recommendations for non-technical stakeholders.

Machine Learning Intern @ ONGC, Dehradun

Tech Stack: Random Forest, XGBoost, Scikit-Learn, K-Means Clustering, Streamlit

Demo: Lithology Classification App

• Built Random Forest and XGBoost models for lithology prediction on 15,000+ geological records, achieving an impressive 87% accuracy via feature engineering.
• Applied K-Means clustering for electro-facies identification, demonstrably improving reservoir zone interpretation efficiency by 25%.
• Developed a Streamlit web application demo for interactive, real-time visualization of geological predictions.
• Presented AI model outputs to geoscience domain experts, incorporating iterative stakeholder feedback over a 3-month engagement.