Elliott Rayne’s Blog.
My notes, frameworks, and focused thinking. I write about what we’re building, what we’re learning, and my predictions for how AI will continue to reshape business.
All Articles
-
The Problem with Shared Credentials in Multi-Agent Systems
Read Article: The Problem with Shared Credentials in Multi-Agent SystemsAn API key can establish that a request came through an authorised connection without explaining which agent was responsible for the operation. I think…
-
Why AI Approvals Need More Than a Yes Button
Read Article: Why AI Approvals Need More Than a Yes ButtonAn approval button can make a workflow look controlled without making the approved action clear. The agent presents a summary. A person clicks yes.…
-
How to Roll Out Enterprise AI Without Handing It the Entire Business
Read Article: How to Roll Out Enterprise AI Without Handing It the Entire BusinessGoing live should not mean granting the system every capability the team eventually wants it to have. I think an enterprise AI rollout works…
-
The Role of Deterministic Controls in Agentic AI
Read Article: The Role of Deterministic Controls in Agentic AISome parts of an AI system should not need interpretation. If an agent is permitted to change one customer's record, the tool should not…
-
Why Your AI Agent Needs a Stop Condition
Read Article: Why Your AI Agent Needs a Stop ConditionAn agent that keeps working is not necessarily making progress. It may be investigating a difficult problem. It may also be repeating a search,…
-
How to Measure the Real Cost of an AI Workflow
Read Article: How to Measure the Real Cost of an AI WorkflowA cheap model can sit inside an expensive operation. I think that is one of the most useful things to remember when discussing AI…
-
Why AI Agents Should Not Treat Every Document as an Instruction
Read Article: Why AI Agents Should Not Treat Every Document as an InstructionAI prompt injection is a useful reminder that reading information and accepting instructions are different responsibilities. An agent may need to read a customer…
-
The Architecture Behind Reliable AI Knowledge Retrieval
Read Article: The Architecture Behind Reliable AI Knowledge RetrievalAI knowledge retrieval can produce an answer that looks well supported while the system has still selected the wrong information. There is a citation.…
-
When One AI Agent Is Better Than a Multi-Agent System
Read Article: When One AI Agent Is Better Than a Multi-Agent SystemA single AI agent can be the right architecture even when the technology makes a larger agent network possible. I think that is an…