3 ms·
"Tell me the most common reasons for tickets" "Find tickets involving feature X" "Find tickets where the customer became angry or agitated" We're doing somet
by nijave 1y ago
"Tell me the most common reasons for tickets"
"Find tickets involving feature X"
"Find tickets where the customer became angry or agitated"
We're doing something similar at work to analyze support cases. We have some structed fields but want to also do some natural language processing on the ticket to extract data that isn't captured in the structured fields.
Think topic extraction and sentiment analysis of ticket text
We're not using MCP but looking into LLM enrichment/feature extraction
- arrowsmith 1y agoNone of those require you to use Cursor. And all of them can be done within a simple environment that doesn't have full read/write access to your entire database.