Emergent Generative AI and LLM technologies can be effective tools for individual tasks. Using them in a repeatable, reproducible manner that is efficient, without sacrificing critical thinking, has proven more elusive. This presentation will outline principles that reduce time and effort without removing the critical 'human-in-the-loop'. These fundamentals can be adapted to any modern LLM platform, and represent an alternative approach to leveraging LLMs.
Presenter: Ward Fisher is a Computer Scientist with a background in Machine Learning. For the last 14 years, he has worked as a Research Software Engineer at NSF Unidata, a community program managed by the University Corporation for Atmospheric Research (UCAR), the managing entity for the National Center for Atmospheric Research (NCAR).