This hands-on course bridges the gap between academic knowledge and industry needs in data science and AI. It revisits core concepts, incorporates modern AI technologies, and uses weekly end-to-end assignments to simulate real-world workflows.

Pre-requisite: BSCS3004 : Deep Learning

Lectures-Tutorials-Practicals-Credits: 4-0-0-4

Week 1: Data science and Python stack refresher (NumPy, Pandas, Matplotlib, Seaborn); Jupyter workflows; Version control with Git.

Week 2: Machine learning with Scikit-learn; Core models (regression, classification, clustering); Pipelines, hyperparameter tuning, evaluation metrics.

Week 3: Deep learning with PyTorch and TensorFlow; Building and training neural networks; Model saving/loading; GPU training.

Week 4: Computer vision fundamentals; Image processing with OpenCV and PIL; Transfer learning with CNNs (ResNet, MobileNet).

Week 5: NLP and LLMs with Hugging Face; Transformer models (BERT, RoBERTa); Tokenization, fine-tuning, and inference.

Week 6: Prompt engineering; LangChain fundamentals (templates, memory); LLM APIs (OpenAI, Gemini, Ollama).

Week 7: Retrieval-Augmented Generation (RAG); Embeddings and semantic search; Vector databases (FAISS, Pinecone, ChromaDB).

Week 8: AI agents and autonomous systems; Tool use and planning; Building agents with LangChain and LangGraph.

Week 9: Data visualization and dashboards; Tools (PowerBI, Tableau, Streamlit, Plotly); Storytelling with data.

Week 10: Time series analysis and forecasting; Models (Prophet, DeepAR); Trend, seasonality, anomaly detection.

Week 11: Big data and cloud platforms (GCP); Data pipelines (BigQuery, Vertex AI); Concepts of Spark, Hadoop, Airflow, deployment.

Week 12: Model explainability (SHAP , LIME); Ethics in AI (privacy, fairness); Mini-project.