At Morosoft Labs, I worked on practical artificial intelligence and machine learning projects using Python. I gained hands-on experience developing machine learning and deep learning models, from preprocessing data through training, testing, and evaluation.
My projects include house price, Titanic survival, and air quality index prediction, as well as handwritten digit recognition and skin disease prediction using transfer learning. I evaluated model performance using metrics including accuracy, precision, recall, RMSE, and R².
I also developed an AI assistant using RAG concepts with Ollama Llama 3.2:3B and Streamlit. I experimented with prompts to compare and improve response quality, using Ollama for local LLM inference and Streamlit for the interactive interface.

