Skip to content
View PiyushKumar74110's full-sized avatar
5 followers · 9 following

Block or report PiyushKumar74110

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
PiyushKumar74110/README.md

Hi , I'm Piyush Kumar

Data Scientist | Machine Learning Engineer | Generative AI Enthusiast

LinkedIn Email

About Me

  • Interested in Machine Learning, Generative AI, and Data Science
  • Currently learning LLMs, RAG, and AI Agents
  • Building AI-powered applications and data-driven solutions
  • Always exploring new technologies

Tech Stack

Programming Languages: Python, SQL

Data Analysis: Data Cleaning, EDA, Hypothesis Testing, Central Limit Theorem, Statistical Modelling, A/B Testing

Data Visualization: Tableau, Matplotlib, Seaborn, Power BI, Excel

Data Wrangling: Pandas, Numpy, Scikit-Learn, SciPy, Excel

AI & LLMs: LangChain, Prompt Engineering, AI Agents, RAG, LLM Evaluation, OpenAI API, Multi-Agent Systems

Cloud Computing: AWS

Tools & Technologies: Git, GitHub, Jupyter Notebook

Projects

Podcast Q & A Bot

YouTube Podcast AI Q&A System is a Retrieval-Augmented Generation (RAG) application that allows users to interact with YouTube videos through natural language questions. The application processes video transcripts, builds semantic search indexes, retrieves relevant content, and generates context-aware answers using Google's Gemini model.

Tech: Python, Google Gemini 2.5 Flash, Sentence Trasformers, Fatser Whisper, LangChain

🔗 Repository: https://github.com/PiyushKumar74110/Podcast-Q-A-Bot

Saaransh-AI

An end-to-end AI system that converts YouTube videos into structured knowledge and enables conversational interaction using Retrieval-Augmented Generation (RAG). The system performs audio extraction, transcription, summarization, insight extraction, and semantic question answering over video content.

Tech: Python, LangChain, RAG pipeline, LLM Orchestration, Mistral AI, OpenAI Whisper, Saravm AI STT model, ChromaDB, HuggingFace (all-MiniLM-L6-v2), yt-dlp, FFmpeg, soundfile, NumPy, NLP and Text Processing

🔗 Repository: https://github.com/PiyushKumar74110/Saaransh-AI

Time Series Batch Prediction

This project is an end-to-end MLOps pipeline designed for time series data analytics and forecasting. It automates data ingestion, preprocessing, model training, evaluation, and batch prediction workflows. The system enables scalable and reproducible analysis of time-dependent data while following MLOps best practices for deployment and monitoring.

Tech: Python, Pandas, Scikit-Learn, XGBoost, Docker

🔗 Repository: https://github.com/PiyushKumar74110/mlops-batch-job

GitHub Stats

Learning, Building, and Growing in AI.

Pinned Loading

  1. Podcast-Q-A-Bot Podcast-Q-A-Bot Public

    YouTube Podcast AI Q&A System is a Retrieval-Augmented Generation (RAG) application that allows users to interact with YouTube videos through natural language questions. The application processes v…

    Python 1

  2. mindguide-ai mindguide-ai Public

    MindGuide AI is a machine learning project that analyzes user behavioral and journal data to understand emotional states and intensity. It combines NLP and structured data to classify emotions usin…

    Jupyter Notebook 1

  3. mlops-batch-job mlops-batch-job Public

    This project is an end-to-end MLOps pipeline built for time series data processing and batch prediction. It automates data ingestion, preprocessing, model training, and inference in a structured wo…

    Python 1

  4. ResearchNet ResearchNet Public

    Python 1

  5. Saaransh-AI Saaransh-AI Public

    An end-to-end AI system that converts YouTube videos into structured knowledge and enables conversational interaction using Retrieval-Augmented Generation (RAG). The system performs audio extractio…

    Python 1

  6. VitalWatch_Pro VitalWatch_Pro Public

    VitalWatch Pro is a real-time health and safety monitoring dashboard built with PyQt5, designed to interface with an Arduino via serial communication. It provides visual feedback and logging for va…

    Python