I am an MBA candidate specialising in Business Analytics at CHRIST (Deemed to be University), with a postgraduate background in Data Science and Artificial Intelligence from MIT xPRO.

My interests sit at the intersection of Applied AI, business analytics, financial services and digital transformation. I enjoy translating business problems into analytical workflows and practical technology solutions.

My work spans retrieval-augmented generation, voice interfaces, data analysis, forecasting and financial research. I am particularly interested in building AI systems that are useful, measurable and grounded in reliable data.

Python · FastAPI · SQLite · Sarvam AI · Groq

An independently developed project for the Paytm Build for India AI Hackathon, Bengaluru (Track 1: Merchant Growth AI). Selected as a finalist.

• Analyses transaction history using SQL-based RFM segmentation to identify lapsed customers and delayed regulars.
• Supports win-back campaigns with deterministic offer and cashback controls.
• Includes an 80/20 treatment-control design to help measure campaign impact.
• Provides voice-led explanations in Kannada, Hindi and English.
• Demonstrated using a synthetic, seeded transaction ledger and a 21-test QA suite.

Independent hackathon project; not an official Paytm product.

Live Demo · Source Code

React · Vite · FastAPI · Deepgram · Groq · edge-tts · BM25

A live, podcast-style voice agent that hosts interactive conversations and retrieves relevant information from user-uploaded context.

• Uses Deepgram for speech transcription and Groq for conversational responses.
• Generates spoken responses using edge-tts.
• Uses BM25 retrieval over uploaded documents to ground conversations in supplied material.
• Includes a React audience interface and an operator console, with a Streamlit dashboard available for alternate deployments.
• Supports session transcripts, document uploads and operator-side diagnostics.

Source repository

Federal Bank · Planning MIS Department · Python · SQL · RAG

Designed and built with the Planning MIS team during my Applied AI internship at Federal Bank. This was a proof of concept, not a production system.

• Developed a retrieval-augmented natural-language-to-SQL workflow across 168 validated Finacle tables from a schema of more than 900 tables.
• Combined query decomposition, domain and temporal routing, BM25 and vector retrieval, Reciprocal Rank Fusion, cross-encoder reranking, graph-based join planning and SQL validation.
• Improved execution accuracy from 25.9% to 83.0% on a 1,000-query benchmark through data-engineering and schema-context improvements.
• Built a reusable knowledge base containing table descriptions, join relationships, decode mappings and query-pattern templates.
• Estimated a potential annual saving of 4,070+ hours; this was a projection, not a realised saving.

Python · Groq · ChromaDB · BM25

A financial research assistant designed to work with Indian company annual reports and RBI circulars.

• Combines dense semantic retrieval and BM25 keyword search using Reciprocal Rank Fusion.
• Uses an agent to select relevant sources and produce answers with supporting citations.
• Submitted to a hackathon at IIT Kharagpur.

Source repository

Python · Pyodide/WebAssembly · WebR · DuckDB-WASM

A browser-based analytical workspace for Python, R and SQL.

• Runs analytical workflows in the browser, keeping data client-side.
• Includes an automatic CSV data profiler and AI-assisted code generation.
• Deployed through a GitHub Actions pipeline.

Source repository

CrewAI · Ollama · LangChain · ChromaDB · Streamlit

A multi-agent platform for enterprise IT governance and pre-implementation project risk review.

• Covers legal and regulatory compliance, data protection and privacy, AML, IT security, governance and SDLC controls.
• Emphasises retrieved evidence, confidence scoring, unsupported-claim flags and an audit trail.
• Includes an MCP server that exposes the agents as tools.

IDEAS, Indian Statistical Institute Kolkata · Python · Random Forest · MLP

Completed as part of the Autumn Internship Program 2026 under the mentorship of Dr. Chandan Biswas.

• Prepared and cleaned historical product-demand data, engineered features and treated outliers.
• Built Random Forest and Multilayer Perceptron models.
• Evaluated model performance using a business penalty metric and documented the analysis in a project report.

UN Volunteers · SDG Research & Data Fellow, UNDP India
November 2026 – December 2026

Selected for a part-time research and data assignment supporting SDG-related work.

IDEAS, Indian Statistical Institute Kolkata · Autumn Internship Program 2026
September 2026 – October 2026

Worked on industrial product demand forecasting, including data preparation, feature engineering, model development and evaluation.

Federal Bank · Applied AI Intern, Planning MIS Department
April 2026 – May 2026

Designed and built SQL Studio, a retrieval-augmented natural-language-to-SQL proof of concept, working with the Planning MIS team.

NIBM Pune · VittNivesh · Equity Research Live Project
May 2025 – July 2025

Completed an eight-week project involving fundamental and technical analysis, financial-results analysis, news and sentiment analysis, and entry/exit timing.

CHRIST (Deemed to be University), Bengaluru
MBA, Business Analytics · 2025–2027

MIT xPRO
Postgraduate Program in Data Science and Artificial Intelligence · September 2025–September 2026
Completed.

St. Joseph's College of Commerce (Autonomous), Bengaluru
BBA, Marketing · 2022–2025

• Programming and data: Python, SQL, Pandas, Git, GitHub
• Analytics and BI: Microsoft Excel, Power BI, exploratory data analysis, forecasting, model evaluation
• AI and machine learning: RAG, LLM applications, agentic retrieval, hybrid search, BM25, vector search, Reciprocal Rank Fusion, prompt engineering, Random Forest, MLP
• Platforms and tools: FastAPI, Streamlit, ChromaDB, Supabase/PostgreSQL, Google Analytics 4
• Business and finance: Financial analysis, equity research, market research, regulatory-document analysis, data quality and stakeholder reporting

• Bloomberg Financial Certifications: Bloomberg Market Concepts, Finance Fundamentals, Spreadsheet Analysis and Environmental, Social and Governance (ESG).
• Oracle Cloud Infrastructure: 2025 Certified AI Foundations Associate (valid June 2026–June 2028).
• Stanford: Code in Place 2025 (certificate issued).

In progress: IIM Bangalore FinTech Certificate Programme for Young Professionals, Batch 2.

• Top 100 Unstoppable B-School Leaders 2026, Unstop.
• Unstop Global Rank 10 among 3.5M+ users; Management Category Rank 4 and CHRIST University Rank 1.
• Finalist, Paytm Build for India AI Hackathon, Bengaluru.
• Second place, Vigil Fraudathon, SPJIMR Mumbai, with HDFC Bank's Vigil Aunty.
• First place, Strategy SIP Off, Daulat Ram College, University of Delhi.
• First place, BoardRoom: The General Management Quiz, IIM Calcutta.
• Third place, Macro Minds Economics Challenge, IIM Calcutta.
• Federal Bank Talent Times Award 2025 for academic achievement.

I have also earned multiple top-three finishes in case competitions, quizzes, article-writing contests and policy challenges hosted by institutions including IIM Calcutta, IIM Udaipur, XLRI, FMS Delhi and SPJIMR.