RAG Engineer

Deep dive into embedding models, similarity metrics, and vector database internals (HNSW, IVF).

• Start hereHNSW vs IVF, when…
• Match each vector index type to its primary characteristic.
• Why is cosine similarity preferred over Euclidean distance for text embeddings?
• What is the primary purpose of embeddings in modern NLP?
• Spot the conceptual error:…
• Match each contrastive loss to its defining mechanism

Master chunking strategies, hybrid retrieval (BM25 + dense), re-ranking, and context management.

• Start hereWhich metric best measures whether a RAG answer is grounded in the retrieved context?
• When should you fine-tune instead of using RAG?
• What does RAG primarily help with in LLM based applications?
• Why combine BM25 with dense embeddings for retrieval?
• Which of these are valid considerations when choosing a chunking strategy for RAG?
• What is the primary purpose of embeddings in modern NLP?

Build production RAG: ingestion pipelines, citation tracking, failure modes, and evaluation.

• Start hereIn LLM serving, what is the primary driver of end to end latency for a generation request?
• Which metric best measures whether a RAG answer is grounded in the retrieved context?
• Design a RAG evaluation harness, what do you measure and how?
• When should you fine-tune instead of using RAG?
• What does RAG primarily help with in LLM based applications?
• Why combine BM25 with dense embeddings for retrieval?

Advanced patterns: multi-hop retrieval, agentic RAG, caching, serving, and observability.

• Start hereWhat is the KV cache in transformer inference?
• In LLM serving, what is the primary driver of end to end latency for a generation request?
• Design a RAG evaluation harness, what do you measure and how?
• Why combine BM25 with dense embeddings for retrieval?
• What is the Model Context Protocol (MCP) and what problem does it solve?
• What distinguishes an AI agent from a plain LLM call?