Obot Learning Center

Building a Retrieval Augmented Generation Chatbot: A Practical Walkthrough

The model behind your support bot finished training before your refund policy was rewritten. It will still answer questions about that policy, fluently and with no signal that its knowledge has expired. A retrieval-augmented generation chatbot closes that gap by fetching the current document at question time and forcing the answer to come from it. […]

Read More

Building Agentic RAG with MCP: A Practical Guide

Meilisearch’s own agentic RAG guide answers the question “what is MCP” with a definition: a prompting technique for structured output design, unrelated to retrieval. It is wrong. MCP is Model Context Protocol, the standard that lets an agent discover and call tools, including retrieval tools, at runtime instead of a developer hardcoding a specific SDK […]

Read More

How to Evaluate RAG Accuracy: Metrics That Actually Matter

A RAG system can score well on the metrics most teams reach for first and still confidently hallucinate in production. BLEU and ROUGE, both inherited from machine translation and summarization, measure how many words in a generated answer overlap with a reference answer. Neither one checks whether the answer’s claims actually trace back to what […]

Read More

RAG Security: Data Poisoning and Privacy Risks

On June 11, 2025, Microsoft patched a vulnerability that didn’t require the victim to click anything. Aim Labs disclosed what researchers later named EchoLeak, CVE-2025-32711, a flaw in Microsoft 365 Copilot that let an attacker exfiltrate sensitive data from a user’s own mailbox and documents using nothing but a single email. No malware. No phishing […]

Read More

Shadow AI Agents & Agent Sprawl: Identity Sprawl vs. Context Sprawl

Two agents, both provisioned through the same approval workflow, both carrying scoped credentials that would pass any identity audit clean, give two different answers to the same question: what was last quarter’s revenue. Neither autonomous agent did anything an identity review would flag. Neither violated a permission boundary. They just don’t agree on what’s true, […]

Read More

GraphRAG: What It Is and When You Actually Need It

Ask a vector search system why two clauses in two different contracts contradict each other, and it will hand you both clauses side by side. It won’t tell you they contradict. Nothing in a nearest-neighbor lookup is built to notice a relationship between two facts, only how similar their embeddings are. GraphRAG, short for graph […]

Read More

Shadow AI Management Best Practices: The Layer Most Frameworks Skip

On May 7, 2026, a compliance officer at CB Financial Services determined that a bank employee’s use of an unauthorized AI tool to process non-public customer records had crossed a threshold. The records included names, Social Security numbers, and dates of birth. Four days later, on May 11, the company filed a Form 8-K under […]

Read More

Non-Human Identity Sprawl in AI Systems

In the average enterprise, non-human identities outnumbered human ones 92 to 1. Entro Security’s H1 2025 research put that ratio at 144 to 1 a year later. This isn’t a static fact about how many service accounts a company happens to run. It’s a growth rate, and most identity management programs were built around a […]

Read More

The Enterprise AI Governance Control Plane: What It Is, What It Actually Does, and Why MCP Changes the Architecture

When enterprise AI deployments move from models to autonomous agents, the infrastructure needed to know what those agents are doing, enforce what they are permitted to do, and prove it to an auditor becomes the AI governance control plane: the centralized layer that unifies agent identity, policy enforcement, and audit trails across every agent, tool […]

Read More

Agentic AI Governance Framework: A 3-Tier Model Enforced at the Point of Action

An agent with broad, unreviewed tool access doesn’t wait for a policy meeting before it acts. It reads the permissions it was given, at whatever scope someone configured months ago, and executes. Nobody audits that scope until something goes wrong, and by the time they do, the question isn’t whether the agent had a written […]

Read More

Claude Cowork Security: What’s Actually Documented, Not What’s Assumed

A researcher connected a single folder to a fresh Cowork session, sent one short message, and watched the agent escape its sandbox. From inside the virtual machine, it read and wrote files across the entire host Mac, credentials included, with no permission prompt at any point. Anthropic reviewed the report and closed it as informative. […]

Read More

What Is an LLM Gateway and How Does It Work?

An LLM gateway is a centralized middleware layer that sits between your LLM applications and various LLM providers: OpenAI, Anthropic, Google, Mistral, and self-hosted models alike. It gives existing code a single endpoint to call instead of juggling LLM APIs with incompatible formats, authentication schemes, and error behaviors. Because LLM deployments increasingly rely on multiple […]

Read More