Convo: Better agentic workflows through pseudocodeThis post is a bit of an update to my now aging INSTRUCT post. Models have come a long way in the last 2 years. My understanding of how…Jan 15Jan 15
Grounding LLM’s — Part 1Hallucinations are currently one of the more troublesome side effects of working with Large Language Models (LLM’s.) If you’re reading…Aug 11, 2024A response icon3Aug 11, 2024A response icon3
Making OpenAI Functions ReliableI’ve seen a lot of developers struggling with getting OpenAI’s new function calling features to work reliably so I thought I’d outline the…Jul 5, 2023A response icon2Jul 5, 2023A response icon2
INSTRUCT: Making LLM’s Do Anything You WantINSTRUCT is a technique for getting LLM’s to execute small programs as part of a prompt. LLM’s are already good at following instruction so…Apr 14, 2023A response icon3Apr 14, 2023A response icon3
