Articles for people who
ship the stack
Rewrites on React, Node.js, TypeScript and AI. A translated body is shown when it exists. If it does not, you see the English text and a note that says so.
Tagged: hallucination
Auditing Order Confirmations an Agent Never Placed: Checking Values Instead of Wording
How to build a claim-consistency evaluator for a support agent that places real orders: partition outcomes by owner, verify identifiers against tool results.
16 min readHow RAG Cuts LLM Hallucinations: Retrieval, Embeddings and Enterprise Use
Why LLMs answer wrong with confidence and how RAG grounds them in current data: the two-part architecture, embeddings, fine-tuning trade-offs and limits.
9 min readTracing Which Hop Introduced a Hallucination With Hadith-Style Provenance
See why LLM errors start at hop one, and how an isnad-inspired chain, narrator registry and decision matrix can give multi-agent pipelines per-hop accountability.
5 min readWhy Smart AI Assistants Still Give Wrong Answers: A Pipeline View
Follow an AI assistant pipeline from training data and context windows to RAG retrieval and generation to see where wrong answers come from and how to reduce them.
7 min readWhen a Model Never Finishes: Silent Failures Hidden From Hallucination Scores
Two checkpoints, two quantizations and two runtimes tested on invented names: one model stays silent for minutes, and a careful grader exposes what scores miss.
11 min readA Portfolio Chatbot Built to Say "I Don't Know" Instead of Guessing
How a portfolio chatbot stays honest: a locked honesty dial, five handwritten notes in the prompt instead of RAG, and a fixed refusal line that doubles as a gap detector.
8 min read
About these articles
Request a 24h estimate
Need the same stack in a production operator layer? Send the brief — estimate within 24 hours.