I completed 30+ end-to-end projects mostly in the domains of NLP, using modern Deep Learning algorithms like LSTMs, GRUs, and various transformer-based models (BERT, RoBERTa, XLM-RoBERTa, etc), as well as LLMs like LLama2 and GPT. I also worked on projects in the areas of Predictive Analytics, Time Series Analysis, Computer Vision, Sentiment Analysis, and Association Rule Learning.Below is a summary of the relevant projects:⚡️ Involved in the development of a Generative AI Model for text-to-image: My role was to develop a system to analyze previous user inputs and recommend optimized prompts to the user. I fine-tuned Meta's LLama2 model for this purpose.⚡️ R&D on Efficient Stable Diffusion Model Loading with Advanced Techniques: I used advanced techniques like model quantization, low-rank adaptation, and knowledge distillation. ⚡️ Real Estate Price Prediction Model: Engineered a machine learning model for price prediction in real estate, utilizing advanced statistical ML algorithms and hyperparameter tuning, with a Flask server backend, and deployed on AWS.⚡️ Market Basket Optimizer Development: I used the Apriori and ECLAT algorithms, aiding in sales growth by analyzing customer purchase patterns.⚡️ Sentiment Analysis for Restaurant Reviews: Executed sentiment analysis on customer reviews for a restaurant using Natural Language Processing (NLTK) and the Naive Bayes algorithm, extracting valuable insights from customer feedback.⚡️ Multi-class Image Classifier: Developed a multi-class image classifier using TensorFlow, Sci-kit Learn, and Keras, with enhanced accuracy through data augmentation and layer dropout techniques.⚡️ Anomaly Detection in Stock Prices: Implemented an LSTM Auto-encoder for anomaly detection in stock price prediction, utilizing TensorFlow and Keras, showcasing advanced skills in time-series analysis and deep learning methodologies.⚡️ Customized LLMs like LLama2, GPT-3.5, & GPT-4 for multiple business use cases.