Priyanshu
2025 — AI Engineer

AI INNOVATION CREATIVE CODING

About Me

I'm an AI Engineer who builds the full spectrum of intelligent systems, from deep learning models(PyTorch, TensorFlow) to advanced Generative AI agents (LangChain, LangGraph). I specialize in deploying these solutions as scalable, production-ready services using Docker and FastAPI.

Priyanshu - Portfolio Image

projects

FastChatBot — RAG-Powered LLM Chatbot

An AI-powered chatbot built using Retrieval-Augmented Generation (RAG) that allows you to query and explore the book "Hands-On Large Language Models" by Jay Alammar.

Combines semantic search with Pinecone vector store and generative answers from a state-of-the-art open-source LLM deployed through Groq. Features a modular, reproducible pipeline using DVC with 5 trackable stages: Document Loader → Text Splitter → Store Index → RAG Inference → Output Generator.

Built with Flask chat UI for elegant real-time communication and uses all-MiniLM-L6-v2 (384-dim) compact embeddings for efficient vector search.

RAGPineconeGroq APIHuggingFaceFlaskDVCPython
FastChatBot RAG-Powered Chatbot Demo

Image Classifier API

Built and deployed a computer vision model that classifies images with 95% accuracy. The model is served via a REST API built with FastAPI and containerized with Docker for scalability.

PyTorchOpenCVFastAPIDocker
Image Classifier API Demo Image

Sales Forecasting Engine

Designed a time-series forecasting model using LSTMs to predict future sales data for an e-commerce client, improving inventory management and reducing overstock by 25%.

TensorFlowLSTMPandasScikit-learn
Sales Forecasting Engine Demo Image

Experience

& Timeline

May 2024 - Aug 2024

AI Research Intern

Tech Innovators Inc.

  • Developed and fine-tuned transformer models for NLP tasks, improving accuracy by 15%.
  • Collaborated with the MLOps team to integrate models into the production pipeline.
Jan 2023 - Apr 2023

Machine Learning Intern

Data Insights Co.

  • Assisted in building a customer churn prediction model using Scikit-learn and XGBoost.
  • Performed data cleaning and feature engineering on large datasets for model training.
Sep 2022 - Dec 2022

Open Source Contributor

Scikit-learn

  • Contributed to the library by fixing bugs and improving documentation for the clustering module.
2022 - Present

Problem Solving Profile

LeetCode, GeeksforGeeks

  • Solved 500+ problems with a focus on Data Structures, Algorithms, and Dynamic Programming.
  • Achieved a top 10% rating in weekly contests, demonstrating strong analytical skills under pressure.
  • Consistently practice medium and hard level problems to sharpen algorithmic thinking.
let's connect