Transcript
Patrick, a Customer Success Manager at Softr, manages many client calls and uses transcripts to extract data and summarize key insights. He uses a sample database designed for an inbound lead meeting tracker where call transcripts are stored.
To automate the processing of these transcripts, you can add a new field in Softr Databases called an AI agent. In this example, the field is named AI Summary to capture actionable items and concise information from each conversation.
[.blog-callout]
While Patrick is manually adding fields here, you can use the AI co-builder to generate entire App templates or database schemas just by describing your needs.
[.blog-callout]
When setting up the AI agent, you can choose from a variety of models from Anthropic and OpenAI. Each model displays the number of AI credits required per use, allowing you to select the version that best fits your needs.
The prompt for the AI agent can utilize dynamic values from other fields within the same record. By selecting the transcript field, you can instruct the AI to summarize the content while specifically including the topic, attendees, company, and next steps.
[.blog-callout]
This is a perfect example of building a internal tool for project management. By using Softr Databases, your team can centralize call logs and AI insights in one secure place.
[.blog-callout]
You can configure the AI agent to trigger manually, when a new record is created, or when an existing record is updated. This ensures that summaries are generated automatically as soon as new call data enters the system.
For existing records, you can manually trigger the AI to fill the field. The agent sends the transcript to the selected model and returns the formatted value with the requested details like topic, company, and next steps.
[.blog-callout]
To further automate your workflow, consider using Softr Workflows. You can trigger native actions within Softr as soon as an AI agent finishes its summary, such as updating a status or notifying a team member.
[.blog-callout]
This workflow helps refine internal processes by using AI agents directly inside of your Softr data. It provides a simple way to transform raw transcript noise into structured, actionable items for your team.
