How do we put large language models to work? Carl and Richard talk to Vishwas Lele about his work using LLMs with his customers. Vishwas talks about focusing on specific data sets for building LLMs and how size matters - things are simple when the source data is small, but as it grows, you need more complex tools to be able to allow the LLM to perform. Lots of cautionary tales and ideas on how to get great results from these new automation tools!