Persist ingestion checkpoints at source boundaries
Track progress per consumer and source ordering boundary, and advance each checkpoint only after the corresponding ingestion work succeeds.
Retrieval
Store a source checksum to skip unchanged documents
Compare a document’s source checksum, metadata, and ingestion profile before repeating extraction, chunking, embedding, and index writes.
Your First Qdrant Collection
Run Qdrant locally, connect with the official .NET client, and create a collection whose vector size and distance match the embedding model.
Separate ingestion health from retrieval readiness
A failed ingestion run should not stop retrieval while an accepted index remains queryable and eligible to serve.
Embeddings in .NET
How to generate text embeddings with Microsoft.Extensions.AI and keep the model, dimensions, comparison metric, and indexing behavior compatible.
Why Retrieval Exists
Why application knowledge does not belong in model weights, what retrieval contributes, and when a prompt, database query, or deterministic tool is simpler.
Measure retrieval misses separately from model failures
Give retrieval its own outcome and telemetry so missing or irrelevant context is not diagnosed as a model problem.
Keep RAG citations tied to retrieved chunks
Citations should point back to the retrieved evidence, not to sources the model guessed after answering.
Agent Framework RAG for Agents: Giving Your Agent the Right Context
How to connect Microsoft Agent Framework agents to private application knowledge without stuffing full documents into the prompt.
Stop RAG Hallucinations with the Short-Circuit Pattern
How to reduce RAG hallucinations by short-circuiting generation when retrieval returns weak evidence, with a simple C# threshold check.