RAG - IoT Digital Twin PLM

Filtered Vector Search: Pre-Filtering, Post-Filtering and HNSW Metadata Filters That Do Not Break Recall

By MPRAUTO MPRAUTO October 10, 2026AI2 Comments

Filtered vector search explained: pre- vs post-filtering, HNSW graph connectivity under filters, ACORN, partitioning and selectivity-aware query planning.

Cohere Embed 5 Pro Explained: Architecture, Benchmarks, Pricing for RAG

By MPRAUTO MPRAUTO October 1, 2026AI1 Comment

Cohere Embed 5 Pro explained: what is confirmed on architecture, retrieval quality vs speed, pricing, and how to migrate a RAG pipeline from older embeddings.

Hybrid Search and Reranking for Production RAG: An ADR

By MPRAUTO MPRAUTO September 30, 2026AI1 Comment

Hybrid BM25 plus dense retrieval with reranking: architecture, fusion methods, latency and cost, and when to skip reranking. A 2026 decision record.

Milvus 3.0 vs 2.6: Lake-Native Vector Search Upgrade Guide 2026

By MPRAUTO MPRAUTO September 24, 2026TechNo Comments

Milvus 3.0 GA (Jul 29, 2026) vs 2.6: External Collections, Storage V3, SINDI sparse index, StructArray, online schema, and the rollback trap to avoid.

pgvector vs Qdrant vs LanceDB: On-Prem RAG Vector Search (2026)

By MPRAUTO MPRAUTO August 13, 2026Kubernetes1 Comment

pgvector, Qdrant, and LanceDB compared for on-prem RAG in 2026: indexing, filtering, latency at scale, and operational overhead.

Agentic RAG Architecture Patterns: When Plain RAG Is Not Enough

By MPRAUTO MPRAUTO August 13, 2026AI1 Comment

Agentic RAG architecture patterns for 2026: planner-executor retrieval loops, tool-augmented search, and when plain RAG breaks down.

LLM Function Calling and Tool Use: A Production Architecture (2026)

By MPRAUTO MPRAUTO July 30, 2026AINo Comments

LLM function calling explained: JSON-schema tool definitions, parallel tool calls, the agent loop, routing, validation and failure handling in a production tool-use architecture.

Matryoshka Embeddings: Adaptive-Dimension Retrieval Architecture (2026)

By MPRAUTO MPRAUTO July 28, 2026AI1 Comment

Matryoshka embeddings explained: how Matryoshka Representation Learning nests multiple dimensions in one vector for adaptive retrieval - coarse-to-fine search, storage cuts, MRL training and failure modes in 2026.

Hybrid Search Architecture: Dense + Sparse Fusion with RRF (2026)

By MPRAUTO MPRAUTO July 27, 2026AINo Comments

Hybrid search architecture explained: fusing BM25 sparse retrieval with dense vector search using Reciprocal Rank Fusion - indexing, scoring, rerankers, latency and failure modes for production RAG in 2026.

Agentic RAG Architecture: Retrieval Inside the Agent Loop (2026)

By MPRAUTO MPRAUTO July 26, 2026AI2 Comments

Agentic RAG architecture explained: moving retrieval inside the agent loop with planners, query rewriting, hybrid search, rerankers and reflection - patterns, evals, cost and failure modes in 2026.