All AI features in Tines are powered by secure access to tenant-scoped large language models (LLMs). There are two core features and our newest product, Workbench, behind AI in Tines.
AI Action
Directly access the LLM at any point in your workflow, with the AI action – one of our 8 core actions. Perfect for summarizing, analyzing, extracting, and making decisions based on data.
Automatic mode in Event Transform
Author data transformations with all the power of code, without the need to read or write it. Simply provide your input and describe the output you need in plain English. Once you approve the behavior suggested by AI, it’s locked into the workflow for consistent execution. Perfect for processing complex, gnarly data structures. Great for deterministic behavior.
Workbench
A natural language chat interface to interact with data and take action on your workflows, privately and securely. Workbench removes the need to switch context between applications. It offers builders a singular view to run and monitor in the context of real-time scenarios or events. Learn more about Workbench here.
Get inspired
As a good rule of thumb, we recommend automatic mode for anything that Python can do well and the AI Action for everything else. Below are some examples for each feature.
AI Action
Summarize an alert or a case
Suggest next steps to an analyst
Explain in plain English
Make a decision to direct automation
Translate between different languages
Build secure AI apps with pages
Automatic mode
Process/convert any data structure
Statistically analyze event data
Map an incoming alert to a record schema
Convert between file formats
Perform advanced cryptography
Add a watermark to an image
Workbench
Send a message in Slack
Look up an employee in BambooHR
Create a ticket in Jira
Get detections in CrowdStrike Falcon
Lock down a device in Jamf
Continue learning about AI in Tines
Article |
Please note: these features aren't supported on self-hosted tenants today.