Lou is currently the principal architect of OmnisCloud, a database-centric SaaS platform serving a variety of vertical market and compute-intensive infrastructure needs in The Cloud, including work in media apps, workflow, document management, medical and bioinformatics, genomics and other areas.
As a developer, he has built and managed projects in Big Pharma, Biotech, Financial Services and the US Military.
Lou connections with Omnis are historic; starting with a DOS version of Omnis3 in the early 1980’s. His company at the time, Essex Systems, had the distinction of being the first Omnis Development Centre in the world. Subsequently, he was a Director of Blyth & Associates which was Omnis’ only foray into direct-to-client professional services, focusing on Omnis’ very largest accounts.
The problem is: AI seems to be everywhere and all at the same time (which can be overwhelming).
This presentation will start with a broad discussion of the current landscape in AI, starting with its foundations in Machine Learning along with an overview of the Frameworks, Tools, Models and Libraries that are available in AI. Lou will discuss some directions AI may be moving in.
At the core of AI is workflows. During the session, we’ll build and demonstrate a standalone AI Workspace environment in n8n. While not quite production-ready, this ‘micro-environment’ will demonstrate all the key elements of an AI implementation over some leading Large Language Models.
We’ll next explore special tools being developed under PostgreSQL to facilitate AI; among these: PgAI and PGvectorscale. With ready access to PostgreSQL from Omnis, Omnis Developers already have AI Superpowers that they did not know they had.
Finally, we’ll look at simple integrations with Omnis, and may even be able to build a few ‘goodies’…
To prepare for this session, it is suggested that each session attendee: