Fakhera Mansuri

Selected for the AI Startups Summer Internship Program. Tasks include reviewing and providing feedback on courses related to building tech and AI startups, completing one course each week and submitting a short review, and identifying and reporting any issues with course quality.

Implemented a Jupyter notebook code to forecast stock prices using Long Short-Term Memory (LSTM) networks with 85% accuracy. Engineered a system predicting survival on the Titanic, leveraging factors like socio-economic status, age, and gender with an accuracy of 78%. Implemented a Handwritten Digit Recognition system achieving 95% accuracy on the MNIST dataset.

Focused on NLP text summarization research using the pointer-generator network. Conducted data preprocessing, implemented neural network architecture in TensorFlow, and fine-tuned hyperparameters. Evaluated model performance using ROUGE and BLEU metrics.

Developed an automated multiple-choice question (MCQ) generation system using Generative AI and advanced NLP techniques, integrating LangChain and OpenAI's language models. Built a user-friendly interface using Streamlit and deployed the application on AWS. Developed a personalized recommendation system for an e-commerce platform using content-based methods, unsupervised learning, NLP, and Deep learning techniques.