AI Loop Engineering
Smolagents vs langgraph.
Guardrails and escalation and judgement
To test the output of your models, you can rely on both deterministic and non-deterministic guardrails (LLM as Judge). Remember that AI outcome is not deterministic so it's safer to rely on deterministic methods when you can, but AI elevates what you can test.
Related: Levels of agentic programming
In this topic
- LLM as JudgeSeedlingfurther reading
- Agentic EngineeringGrowingfurther reading
Reviewing code is a skill that won't go away. No shortcut to it, get good the old school way. Take the time to write the project description and code structure yourself, it…
← Comes after
- Levels of agentic programmingGrowingfurther reading
Use AI for code reviews when the risk is low.