Synchronizing Archive
Synchronizing Archive
Data Scientist and open-source contributor focused on building dependable AI systems.
Transforming raw data into game-changing insights.
Driven by curiosity and powered by data, I am a Data Science enthusiast dedicated to building dependable systems using open-source tools. Analytics fuels my innovation, and data crafts my stories.
End-to-End MLOps.
Scalable logic.
Professional Implementation
Custom RAG-based chatbots tailored for enterprise knowledge bases and workflow automation.
Know MoreFull-stack web solutions focusing on high performance, scalability, and modern user experience.
Know MoreComprehensive statistical data analysis and interpretation using SPSS for academic and corporate research.
Know MoreEnd-to-End implementation of machine learning models from data cleaning to model deployment.
Know MoreScroll to navigate my professional milestones
AIMIT, Mangalore
SGPA: 9.35 | CGPA: 9.143
SDM Degree College, Ujire
SGPA: 8.81 | CGPA: 8.17
FastAPI / GCP / Docker / GitHub workflows / ML Pipeline
Built an end-to-end MLOps pipeline for predicting employee attrition using automated data processing, model training, and deployment workflows, enabling HR teams to identify high-risk employees and make data-driven retention decisions with scalable, production-ready ML systems.
FastAPI / Gemini / Docker / ChromaDB / LLM Ops/ RAG Pipeline
Designed a production-ready LLM Ops RAG pipeline using FastAPI, Gemini, ChromaDB, and Docker for scalable, context-aware AI applications. Implemented containerized deployment and efficient retrieval workflows to ensure high accuracy, low latency, and reliable LLM lifecycle management.
FastAPI / Groq / Docker
A personalized recommendation engine that leverages semantic search and Large Language Models to interpret user queries about movies. Built with FastAPI for the backend, Groq for high-speed inference, and Docker for containerization, this chatbot goes beyond keyword matching to understand the context and mood of user requests, providing tailored cinema suggestions from massive datasets.
Collaborative Filtering
A sophisticated machine learning system designed to suggest books based on user reading history and preferences. Utilizing collaborative filtering techniques, it analyzes patterns in user behavior to identify similar readers and recommend titles they enjoyed. The system addresses the cold-start problem and scales efficiently with growing user bases.
Time-Series / Python
An analytical tool focused on forecasting stock market fluctuations using historical time-series data. By implementing Moving Averages and other statistical methods in Python, this project visualizes trends and attempts to predict future price movements. It serves as a foundational tool for understanding technical analysis and algorithmic trading concepts.
Efficiently read 24+ file formats including PDFs, YAML, and images with built-in logging and custom exceptions.
Building dependable systems through applied machine learning.