Inspiration
Data readiness is the #1 roadblock in enterprise data and AI initiatives.
What it does
TRUIFY enables organizations to use the 90% of data that is not (yet) ready to include in the enterprise data corpus. We make AI more accurate, more responsible and more compliant using expert agents.
How we built it
TRUIFY is a python-based multi-agentic framework. The community edition offers a Streamlit front-end enabling no-code data compliance and reporting. The enterprise edition (new this weekend!) enables Snowflake read/write (data) and Cortex model selection (genAI) all hosted natively within Snowflake (security/ease-of-deployment)
Challenges we ran into
Coding LLMs are not familiar enough with Streamlit, especially how to manage session state variables. Documentation on how to implement Cortex model usage from within Snowflake Container Services is sparse.
Accomplishments that we're proud of
We went from a an open source project to an enterprise-ready solution atop Snowflake, thanks to the exceptional functionality afforded by Snowflake-native apps.
What we learned
Snowflake build and deployment processes via SQL- very slick!
What's next for TRUIFY.AI
We would like to connect with Snowflake sellers. We believe TRUIFY has the potential to not only increase AI accuracy and responsibility, but 10x data stored in Snowflake (consumption).
Log in or sign up for Devpost to join the conversation.