Elliott Rayne’s Blog.

AI observability can give a company an impressive amount of information about its models while leaving it surprisingly uncertain about the work those models…

AI workflow selection is often treated as a search for the most impressive thing the technology can do. I think that is usually the…

An AI workflow and an autonomous agent can produce a similar-looking experience for the person using them. A request arrives. The system interprets it.…

An AI tool failure can leave the agent with a question that a better explanation cannot resolve. Did the operation happen? Imagine an agent…

A source of truth becomes important the moment an AI agent has to do something with the information it finds. For a general conversation,…

An AI handoff can look successful on a diagram and still be a poor experience for everyone involved. The agent reaches a difficult situation.…

AI agent ownership becomes much more interesting after the deployment is finished. Before that point, responsibility normally looks reasonably clear. There is a project.…

AI memory sounds like a problem that should become easier as models improve. Give the agent a larger context window. Let it read more…

AI agent testing becomes a very different problem once the system can do more than answer a question. A response can sound excellent. The…