03.2025
March 2025 release notes.
The March 2025 PolyAI Agent Studio release includes enhancements to dashboards, voice management, knowledge base performance, and real-time collaboration features.
Expand the items for details:
A redesigned standard dashboard now includes additional metrics and improved performance.
What’s new:
- Containment rate, total calls, SMS stats, and function call metrics now available at a glance.
- Hourly refresh cycle brings data closer to real time.
- Dark header and improved layout boost usability and visual clarity.
The new safety dashboard mirrors the updated overview dashboard experience.
What’s improved:
- Faster load times and more responsive data rendering.
- Hourly refreshes ensure up-to-date monitoring.
- Unified styling with clearer visual hierarchy for incident triage.
You can now click matched knowledge base topics in conversation review to open them in a new browser tab.
This allows you to explore topic content without disrupting the review flow, making it easier to cross-reference and validate responses.
Agent Studio now supports real-time collaboration and edit tracking across your team.
Multi-user refresh view in action
How it works:
- See who else is editing a draft in real time.
- Receive notifications when changes are saved by others.
- Copy unsaved edits before refreshing to avoid overwriting.
- View full change logs using the history panel.
Reviewed conversations now include the agent version that was active at the time.
Why it matters:
- Ensures reviewers can account for environment-specific behavior.
- Useful for debugging across staging vs. production versions.
- Supports clearer QA documentation and internal audits.
You can now upload new voices directly in Agent Studio without needing backend intervention.
The voice selection interface has been redesigned with filter controls to streamline voice browsing.
Available filters:
- Language
- Accent
- Gender
SMS messages are now triggered inline during function execution, rather than being queued afterward.
Why it matters:
- Enables immediate error handling for failed sends.
- Prevents agents from falsely confirming delivery before it’s complete.
- Supports tighter runtime logic and better user feedback.
For questions or additional support, contact us at platform-support@poly-ai.com.