Shrey Mishra
Pursuing a Bachelor's degree in Computer Science, expected 2027.
Specializing into the depths of Machine Learning
Experience
-
Artificial Intelligence Intern | BetaZen InfoTech, Remote
Feb. 2026 - Apr. 2026Developed an automated system to generate questions and answers using reference PDF files for Classes 5-8, covering all textbooks and lessons under the West Bengal Bengali Board curriculum.
Projects
-
House Prices Prediction-Kaggle Notebook
Built a machine learning model to predict housing prices using regression techniques and feature engineering. Submitted as a Kaggle notebook for the House Prices: Advanced Regression Techniques competition. Focused on data cleaning, analysis, and model evaluation to improve prediction accuracy.
-
Personalized Recommender System with Unsupervised User Segmentation-Kaggle Notebook
This project is an end-to-end data pipeline that combines unsupervised learning and recommendation algorithms to analyze user behavior and provide personalized content suggestions using the MovieLens-100K dataset.
-
RPS Classifier (CNN + Transfer Learning)-Kaggle Notebook
Built an image classification model to recognize rock, paper, and scissors hand gestures using a baseline CNN and MobileNetV2 transfer learning with fine-tuning, achieving ~99%+ validation accuracy.
-
IMDB Sentiment Analysis (Baseline vs LSTM vs GRU)-Kaggle Notebook
Built an NLP sentiment classifier on IMDB reviews using embeddings and sequence models (LSTM, GRU), and compared them using accuracy/loss trends and confusion matrices.
-
DocuRAG
Built a containerized document Q&A system using Retrieval Augmented Generation (RAG) with FastAPI, Google Gemini API, and ChromaDB, enabling users to upload documents and receive AI-powered answers with streaming responses; fully deployed with Docker Compose for seamless scalability.
-
HireBot
Developed an AI-powered hiring assistant using LLaMA 3.3 via Groq API with Streamlit frontend, implementing structured 5-stage interview pipelines and dynamic prompt engineering for context-aware technical Q&A generation. Integrated DistilBERT sentiment analysis with real-time candidate profiling, session analytics, and GDPR-compliant data export.