The success of agentic applications using LLMs depends largely on the ability to properly manage the context - the collection of prompts, tools, history, memory, and RAG-indexed content. When you take all of these elements into account, you are able to get the most out of the LLM while avoiding hallucination, conversational drift, and relevance issues. Topics to be covered include: dynamic LLM selection, automated prompt development, context compression, tool decoration, RAG optimization, and LLM-as-a-judge quality assessment of LLM responses. Each of these smaller parts together build...
Joshua Powers
Technical Director for AI/ML
Dev Technology
Joshua Powers is the Technical Director for AI/ML at Dev Technology Group. Josh has 30 years of experience in researching and applying artificial intelligence technologies to a wide variety of business and public sector challenges. Prior to Dev Tech, he was VP of Data Analytics at WorldAware where he developed a large-scale open source intelligence exploitation platform to automatically address geospatially and topically diverse intelligence requirements. Josh has been principal investigator on research topics for AFRL, DARPA, and IARPA. He has been awarded patents in semantic search, document clustering, and concept learning. Josh has been an invited speaker and panelist at numerous conferences, including KM World, RSA, and WorldComp. Josh graduated from Stanford with a degree in Artificial Intelligence in 1994.