How to build an AI agent that actually works

"The company's code review 'happens in a workflow' with 'two agentic loops' embedded at specific points where reasoning is actually needed. Not an autonomous AI roaming free. For CodeRabbit, their agent is a workflow with intelligence inserted where it counts."

"Loker describes CodeRabbit's architecture as 'a workflow with models chosen at various stages... with agentic loops using other model choices.' The system doesn't start with a large language model (LLM) and hope. It runs a deterministic pipeline that fetches the diff, builds the code graph, runs static analysis, identifies changed files, determines review scope, and then inserts agentic steps where judgment is actually needed."

"'There are some things that we know are very important so we run them anyway,' Loker says. 'The code graph analysis, import graph analysis, having this static analysis tool information there, the diff, and some of the file-level information.' This base context gets assembled deterministically before any reasoning model is invoked."

Agents succeed in production when embedded within workflows rather than deployed as autonomous systems. CodeRabbit's code review system exemplifies this approach, using deterministic pipelines to gather context before inserting agentic reasoning loops at critical junctures. The architecture prioritizes workflow design over model selection, assembling essential information through static analysis, code graphs, and diff analysis before invoking reasoning models. This hybrid approach combines deterministic processes for reliable operations with targeted AI reasoning where actual judgment is required. Eight core principles and a ten-point checklist guide agent development across various applications beyond code review.

Read at InfoWorld