Round 1: Breadth Assessment This round evaluated the width of my knowledge across the Data Science spectrum. The structure was: Personal introduction Project walkthrough (one detailed project explanation) Technical questions spanning: Machine Learning: Data preprocessing and model evaluation Deep Learning: Optimizers and Gradient Descent Generative AI: RAG (Retrieval-Augmented Generation) and LLMs Coding problems: Printing series patterns and list/dictionary comprehension Difficulty level: Easy to moderate. Round 2: Deep Dive Technical Round This round went significantly deeper into specialized topics: Sentence transformers and their applications Benchmarking and evaluation methodologies RAG architecture and implementation Evaluation frameworks (RAGAs, DSPy) Transformer architecture fundamentals Advanced concepts: Training different word embeddings, contextual awareness, positional encoding
