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.
Rag
Keep RAG citations tied to retrieved chunks
Citations should point back to the retrieved evidence, not to sources the model guessed after answering.
Keep vector search filters separate from semantic ranking
Use filters for eligibility and ranking for relevance instead of blending both concerns.
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.
RAG with EF Core and pgvector
How to build RAG retrieval in .NET by storing embeddings in PostgreSQL with pgvector and querying them through EF Core.
Controlling Token Growth with Chat Reducers
How to control token growth in Microsoft Agent Framework with message-count and summarizing chat reducers, including setup, tradeoffs, and when each approach fits.
Context Compression in .NET
Reduce token cost and latency in .NET by compressing RAG context with a cheap summarizer model or an IChatClient middleware pipeline.
Indirect Prompt Injection Is a Trust Boundary Problem
Indirect prompt injection is a trust-boundary failure; treat retrieved content as untrusted data, isolate it from instructions, and validate actions before execution.
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.
RAG Is a Data Problem Before It’s a Prompt Problem
Why stale documents, weak chunking, and thin metadata usually break RAG before prompt tuning does.