work.md

Projects

These are some of the projects that I have completed uptil now.

March 2026 - Present

PiiScrub

PiiScrub is a blazing-fast, lightweight Python library and CLI tool designed to scrub Personally Identifiable Information (PII) from datasets for LLM training and RAG pipelines. It supports multi-core parallel processing, synthetic data generation, and compliance auditing out of the box with zero external dependencies. Designed for teams working on LLM fine-tuning and RAG pipelines who need compliance-safe training data without external service dependencies.

Technology UsedPython
January 2025 - Present

Django Finance Tracker

TrackMyRupee: A privacy-first AI finance tracker built for Indian salaried professionals. Features include natural language expense entry, AI-powered spending insights and forecasting, net worth tracking, and recurring expense management. No bank login or SMS access required. Open-source.

Technology UsedPythonDjangoAI / LLMsOpen Source
January 2019 - Present

pyresparser (Resume Parser)

pyresparser: A widely-used resume extraction library that parses unstructured resume data into structured formats using SpaCy and custom NER rule-based matching. Integrated into HR-tech products worldwide.

Downloads: 450,000+ on GitHub/PyPISource Code: GithubKnow More: Click here..
Technology UsedPythonDjangoNLTKSpaCyNLP

Professional Highlights (Client Work)

Text-to-SQL GenAI Agent (Emergys Solutions, 2023)

Serverless backend for a natural language to SQL interface using FastAPI, LLMs, and enterprise security patterns (JWT, rate limiting). Drove $300K+ in client revenue.

PythonFastAPIAzureLLMsServerless

Autonomous Supplier Discovery Agent (Emergys Solutions, 2023–2024)

Azure AI Agents that autonomously scrape, parse, and verify supplier capabilities from unstructured web sources for a global supply chain platform.

PythonAzure AILLMsWeb Scraping

Marketing Mix Modeling Platform (Emergys Solutions, 2022–2023)

Probabilistic MMM framework using Meta's Robyn to quantify channel-wise ROI across digital and offline media. Outputs budget allocation strategies for C-suite decision-making.

PythonR (Robyn)AzureAirflow

Also view some of my mini projects