Screening Background, projects, motivation. Basic Python (lists, dicts, file handling). Quick debugging task. Data Foundations Pandas: cleaning, groupby, missing data. Matplotlib: plots, subplots, customization. Hands-on: load → process → visualize dataset. ML & PyTorch PyTorch basics: tensors, autograd, training loop. Build & explain a simple neural net. Optimizers, overfitting, regularization. Transformers (75 mins) Attention mechanism & architecture. Hugging Face: tokenizers, pretrained models. Case study: fine-tune BERT for classification. RAG Design Retriever + generator concept. Pipeline with embeddings & vector DB. Use case: Q&A over company docs. Discuss scalability, hallucinations, retrieval quality. Final Round End-to-end design: RAG assistant for financial documents. Covers preprocessing, embeddings, retriever, fine-tuning, evaluation. Behavioral: teamwork, ownership, learning mindset.