Photo by David Clode on Unsplash A lot of the time you can ask a large language model (LLM) to perform a task and it will give you a decent answer. For example here’s the start of a poem about a pirate doing a code review: In the glow of a lantern, low and dim, A pirate stares at the screen’s rim. “Arr, mateys, this code be a mess, Spaghetti lines in a deep distress. However, for more complex tasks or tasks involving multiple steps LLMs are a lot less reliable. Anyone who’s tried to incorporate ...