FOLIOAIthon
In our user case exercises today, we will adopting some technique and ideas from hackathons1. Original held as a start-up activity, hackathons are made up of teams in a competition that attempt to solve or at least speed up software development of specific problems in a particular industry.
Team Use-Cases
After each topic area or section, we will break up into small teams, with each team working on one of the these use-cases.
Each break-out session will have a theme based on the topics covered to brainstorm, write, and experiment as we improve the wiki pages for the use-cases. Each team will write Github issues in either edge-ai or the ai-workflows based on the wiki. Please be verbose in your Github ticket descriptions and include example prompts, expected outputs, and even screen-shots from FOLIO.
For those interested in actual hacking code, I suggest coding stub edge-ai functions and/or Airflow DAGS and create pull-requests referencing the issues your group creates.
I encourage each to group to use generative AI in these tasks!
Although we won't be declaring a winner of the hackathon today, informal kudos will go to the group with the most Github activity in either repository.
My Use Case: Automated Metadata Generation Enrichment
I'll go first and illustrate some of the possibilities of what can be done today.
I've been working on the Automated Metadata Generation Enrichment over the past year and have created both edge-ai API calls and Airflow DAGs to implement the requirements in this use-case.