Katakam Prashanth

Data Scientist specializing in EDA, SQL Analytics, and dashboard development. Passionate about exploring diverse data domains, applying machine learning models, and deploying solutions on Streamlit and Hugging Face platforms.

GitHub LinkedIn Email

Projects

FIFA 20 Player Analysis

EDA across ~18k players—ratings, wages, clubs, countries, role patterns; clean viz & insights.

EDAData CleaningAggregationVisualizationPandasNumPyMatplotlibSeaborn

Code · Write-up

MSME RAG Chatbot

Profile → scheme matching with explainable eligibility using MongoDB + HF + Streamlit.

RAGPromptingEvaluationPythonHugging FaceMongoDBStreamlit

Code · Write-up

Skills & Tools

Skills

AggregationData CleaningEDAEvaluationPromptingRAGVisualization

Tools

Hugging FaceMatplotlibMongoDBNumPyPandasPythonSeabornStreamlit

Papers

JNANA: A Short-Answer QA Dataset with Multi-Auditor Validation

Katakam Prashanth, et al. · Preprint (2025)

Released a rigorously audited short-answer dataset with a 5-reviewer workflow, detailed rubric, and leaderboard; includes baseline models and analysis.

PDF · Zenodo · DOI · Code · Poster · Slides

Certificates

Vector Search Fundamentals

MongoDB · Issued June 13, 2025

Learned how to leverage MongoDB Atlas Vector Search to build intelligent, AI-powered search experiences for applications. Explore indexing, embeddings, and retrieval strategies.

Link

Blog

Explore My Technical Blog

Dive into technical articles, tutorials, and insights on MongoDB, SQL, Data Science, and Machine Learning

View All Blog Posts →

Contact

Open to roles in ML/AI, RAG systems, and practical CV. For collaborations or consulting, email me at prashanth01071995@gmail.com.

GitHub LinkedIn Email