← DP-800 study guide
Domain 3 – Implement AI Capabilities in Database Solutions (25–30%)
6 lessons, each ending with a practice scenario (Pro)
Log in to save your progress.
- ✓3.1 — Evaluate and Manage External ModelsThe database calls AI models through external model objects: a named, secured definition of where a model endpoint lives, how to authenticate, and what it's for. First you choose a suitable model, then you register it.Scenario: Pro
- ✓3.2 — Embeddings: Columns, Chunking, Generation and MaintenanceAn embedding is a vector that captures meaning, so similar texts get nearby vectors. Good semantic search depends on what you embed, how you split it, and how you keep it current.Scenario: Pro
- ✓3.3 — Choose Full-Text, Semantic Vector or Hybrid Search, and Implement Full-Text Search"Intelligent search" isn't always vector search. Choose the technique from how users search and what must match.Scenario: Pro
- ✓3.4 — Vector Data, Vector Functions, Indexes and ANN vs ENN SearchSQL Server 2025, Azure SQL Database, Azure SQL Managed Instance and SQL database in Fabric store embeddings in a native vector type and search them with exact (ENN, also called kNN) or approximate (ANN) nearest-neighbor search.Scenario: Pro
- ✓3.5 — Implement Hybrid Search with Reciprocal Rank Fusion, and Evaluate ItHybrid search runs a keyword search (full-text) and a semantic search (vector) for the same query, then merges the two ranked lists. Reciprocal rank fusion (RRF) is the standard way to merge them, because it only uses ranks, which avoids comparing incompatible scores.Scenario: Pro
- ✓3.6 — Design and Implement Retrieval-Augmented Generation (RAG)RAG grounds a language model's answer in your data: retrieve relevant rows, pass them to the model as context, and return the answer. In SQL, the whole flow can run in T-SQL with spinvokeexternalrestendpoint.Scenario: Pro
Domain recapQuick overview of all lessons
- Evaluate external models and register them with CREATE EXTERNAL MODEL
- Design embeddings: which columns, how to chunk, how to generate them, and how to keep them current
- Choose between full-text, semantic vector and hybrid search, and implement full-text search
- Use the vector type, vector functions and DiskANN vector indexes, and choose ANN or ENN
- Implement hybrid search with reciprocal rank fusion and evaluate search quality
- Build retrieval-augmented generation with sp_invoke_external_rest_endpoint