Overview: The GuffGaaf project is a workout-based chatbot built using NLP, an LSTM model, and word2vec embeddings. It can parse natural-language queries about exercises, suggest routines tailored to user goals (strength, hypertrophy, endurance), recommend progressions based on available equipment, and offer quick form and safety tips.

The conversational model was trained on a curated dataset of exercise descriptions, Q&A pairs, and fitness guidelines, with data augmentation and synonym expansion to improve robustness.

Idea: The idea emerged from my passion for fitness and also my interest toward creating a chatbot during my course study of NLP.

Challenges: Some of the challenges i faced during this project was during the training of the model, including the manual development of 500+ intents to achieve a proper model.

Technologies Used: Python, LSTM, word2vec embeddings.