01 / Inside the engagement
Situation
The client was developing a natural-language way to explore its analytics data. The application needed to connect business questions to the right data, preserve useful work between sessions and turn results into views people could use.
Context
- Business questions
- Data definitions
- Analytics data
UTIS contribution
- Semantic models
- Cortex Analyst application
- Reusable queries
Simplified workflow. The case text describes the contribution, delivery stage and evidence.
Explore the full conceptual diagram
02 / Inside the engagement
What UTIS did
- Developed a Streamlit application using Snowflake Cortex Analyst, with semantic models that map business terms to the underlying data.
- Implemented saved queries, persistent sessions and chart presentation improvements in response to client feedback.
- Reviewed prompts and outputs, tested edge cases and documented issues and follow-up changes.
- Prepared deployment and handover documentation, working with the client on demonstrations and the next set of enhancements.
03 / Inside the engagement
Results
- The application and Snowflake integration were delivered, with deployment and QA recorded in the engineering work history.
- The client separately confirmed UTIS had developed the Snowflake integration and Cortex AI work.
- Saved queries, session persistence and documented handover supported continued development of the application.
Why it matters to you
A useful AI interface needs more than a chat box. The data definitions, query behavior, session handling and testing all need attention before a team can evaluate it in its own business context.
Stack
Evidence basis: client confirmation of UTIS development, supported by delivery, deployment and QA records. These records establish the work delivered; they do not establish adoption, answer accuracy or financial return.