Below you will find pages that utilize the taxonomy term “Projects”
Maestro
If you’re looking for a general-purpose agent framework, you’ll probably want something like CrewAI or LangChain. But if you’re building robust software and want a system that reflects actual engineering best practices out of the box—welcome to Maestro.
Maestro is a highly opinionated multi-agent orchestration system written in Go. It’s designed from the ground up to manage the full development lifecycle using two core agent roles: a singleton Architect and any number of Coders.
Clearsky
The Clearsky project is an AI-assisted social media client for Bluesky. Effectively, it’s a personal content moderator and social media assistant.
The project aims to use generative AI to solve a number of issues with the current social media environment:
-
Overload. The sheer volume of posts in unmanagable. Clearsky uses summarization to help.
-
Toxicity. Clearsky summaries are meant to be neutral in tone and provide warnings where the underlying content is profane, incendiary, violent, or in other ways objectionable.
CARES
The CARES project stands for Counseling, Assistance, Resources, Empathy, and Support and is an ongoing research project being conducted with data scientist Sherry Towers and anti-violence pioneer Gary Slutkin. The goal is to see whether the current state of large language models (LLMs) is sufficient to provide support to victims or bystanders of bullying behaviors in schools.
The project remains very much at the research stage as the PIs investigate various possible interventions before beginning any trials.
IBIS
IBIS stands for Injured Bird Information System. The project began when one of our principals, Stacy, and her husband Ed helped rescue a mother duck and 12 ducklings who had gotten disoriented and stuck in the middle of a highway.
Not knowing what to do to help the ducks, they called a local bird rescue who got the ducks to safety. Soon after, Stacy and Ed discovered that the texting system the organization was using to organize its volunteers meant that every volunteer got every notification independent of their specialty, equipment, available, or location. IBIS was built to solve this.