Pranjal Goyal

Aspiring data scientist with hands-on experience at IIT Kharagpur, proficient in Python and machine learning. Demonstrated strong problem-solving abilities and effective teamwork in projects, leveraging tools like TensorFlow and Pandas to drive impactful results. Committed to continuous learning and adaptability in fast-paced environments.

Smart Medical Assistant, Generative AI, Agentic AI, 02/01/25, 03/01/25, Built a medical recommendation system using LangChain, LangGraph, and OpenAI LLMs., Orchestrated a multi-node LangGraph workflow for symptom parsing, prediction, and doctor recommendations., Integrated Tavily Search API to fetch and filter real-time doctor information based on user location., Developed custom LLM prompts for generating specialist doctor suggestions from emergency action plans., Optimized backend workflows by modularizing API calls, and managing environment variables. Stock Price Prediction, Time Series Analysis, Deep Learning, 11/01/24, 12/01/24, Developed a 2-layer LSTM model in Keras, achieving 3.97% RMSE across 6 S&P 500 companies., Processed and scaled 30,000+ historical data points from 2013-2018 using Pandas and MinMaxScaler., Visualized price trends and trading volume patterns for Apple, Microsoft, Meta, Amazon, NVIDIA, and Alphabet., Extracted and prepared training and testing data, achieving an average RMSE of 3.975% across all companies. State Farm Distracted Driver Detection, Computer Vision, Deep Learning, 09/01/24, 10/01/24, Achieved 94% validation accuracy on State Farm's distracted driver dataset using a 3.3M+ parameter CNN., Handled and normalized 20,000+ images using OpenCV, with a custom preprocessing pipeline., Designed a Convolutional Neural Network (CNN) in Keras with 3 Conv2D and 2 Dense layers., Reduced overfitting by 80% dropout and optimized training through batch processing and early stopping. Sentiment Analysis, NLP, Deep Learning, 04/01/24, 05/01/24, Devised and optimized multiple RNN models for sentiment analysis on the IMDB movie reviews dataset., Preprocessed IMDB dataset, using padding sequences to a uniform length of 800, handling 50,000 movie reviews., Utilized Embedding layers to transform input words into 64-D vectors, optimizing RNN model performance., Attained high accuracy on test data: SimpleRNN (70.8%), GRU (85.6%), LSTM (84%), and BiLSTM (85%).

• Excelled in the JEE Mains 2021 by ranking in the top 0.8% out of nearly 1.1 million candidates.
• Achieved a rank within the top 3% out of 250,000 students appearing for JEE Advanced in 2021.
• Secured State Rank 2 with a score of 99% in the CBSE Secondary School Examination (Class X).

Programming and Data Structures (Theory & Lab), Probability and Statistics, Linear Algebra, Numerical and Complex Analysis, Advanced Calculus, Economics, Optimization Techniques, File Organization and Database Systems, Computational Methods and Mining Instruments, Application of AI/ML/DL, Data Science and Machine Learning - Coding Ninjas, Machine Learning Specialization - Coursera

C/C++, DSA, Python, Machine Learning, Deep Learning, Data Analytics, MySQL, NumPy, pandas, Matplotlib, Seaborn, FastAPI, Selenium, Scikit-learn, TensorFlow, Keras, OpenCV, Langchain, LangGraph, HuggingFace, MongoDB, ChromaDB, Pinecone, VSCode, Jupyter Lab, Jupyter Notebook, Tableau, DB Browser, Google Colab, Microsoft Excel, Time Management, Team Work, Problem Solving, Communication, Adaptability, Critical Thinking