A decompose-then-verify pipeline that checks each claim in an AI output against its source document, and an honest look at what the RAGTruth numbers actually say about it.
How to use your existing Notion kanban board as the cockpit to manage your running AI agents, demonstrating human intervention and agent memory across runs.
Fine-tuned MiniLM exported to ONNX and run inside an existing Spring Boot app. No vector DB, no Bedrock per query, no GPU. The post I needed when I started.