Karunya Muddana
AI/ML & Software Engineering | B.Tech CSE @ GITAM
Hyderabad, India
B.Tech CSE student interested in AI/ML, LLMs, and building practical software. I work with Python, React, FastAPI, LLMs, RAG, and MLOps, and enjoy building AI-powered projects and experimenting with new technologies.
AI Engineer (Contract)
Healthcare certification project, Germany (Remote)
- Building an AI system that reviews certification applications by checking uploaded documents against what applicants entered
- Caught and fixed a bug where valid scanned certificates were read as expired, which would have sent good applications back
- Fixed a rule that rejected whole applications over one missing file, so the file is now requested instead
- Added a second, independent reading of each document that flags disagreements for human review, with a 33-case test set
Open Source Contributor
IPython, git-cola, icalendar, scikit-learn
- Got fixes merged into IPython (image display) and git-cola (error reporting) within two weeks
- Have further fixes in review at icalendar and scikit-learn
- Build and maintain open-source projects including ExcelMCP, Native and engraph
Freelance Developer
Chemical manufacturer, Hyderabad
- Designed the spreadsheet system the operations team ran their day-to-day work on
- Built an AI assistant that let staff ask questions about their Excel data in plain English, used daily until the company moved to an ERP
- Open-sourced the engine behind it as ExcelMCP
ExcelMCP — MCP Server for Live Excel Data
Open-source MCP server for querying live Excel workbooks through Microsoft Graph. Uses semantic routing to identify the right workbook and provides answers grounded in live spreadsheet data.
Tareekh — Practice Memory for Indian Litigators
AI-powered legal memory system that turns diary photos, notes, and court order sheets into structured memories for each hearing, then answers questions with citations back to the original source. Led the team and built the AI agent and MCP server.
Native — Multi-Provider Agent Runtime
Built for Smart India Hackathon 2026. AI agent that reads documents, analyses data and writes Word, Excel and PowerPoint reports, switching between AI providers and a local model so long tasks survive outages.
Ames Housing — ML Price Prediction Pipeline
Machine learning pipeline for house-price prediction, benchmarking 11 models and tuning XGBoost to achieve a 0.1002 log-RMSE and $13,082 MAE. Uses a scikit-learn Pipeline to prevent validation-data leakage.
B.Tech Computer Science AI
GITAM · 9.2
Intermediate / Class 11–12 (MPC)
Deeksha Junior College · 9.15
Class 10 / CBSE
Kennedy High – The Global School
Attendee · Oct 24, 2026