More powerful model

This is the deployment of the more powerful model we had to rollback in October, now hidden behind a Feature Toggle that will allow us to roll it out customer per customer. For now, its activation is only possible for new customers (and it is activated by default for them). A migration for current customers will be coming soon.

It comes with a new way of generating criteria: they are now enriched with AI, informing users on how their prompts were interpreted, and allowing them to make adjustments if needed. It should help our customers design powerful criteria, and result in more accurate evaluations.

This release has been rolled back. We'll keep you posted when the new model is available again on prod

More powerful model

We have deployed a new model for QM evaluations: it is more powerful, more accurate, and faster than the previous one, and should improve the performance of our evaluations.

It comes with a new way of generating criteria: they are now enriched with AI, informing users on how their prompts were interpreted, and allowing them to make adjustments if needed. It should help our customers design powerful criteria, and result in more accurate evaluations.

Bugfixes

• We fixed an issue with the audio waveforms that were not displayed for mono audios and voicemails

Mono files in Quality Monitoring

It is now possible for a customer to send mono files with multiple speakers over the QM. This option can be enabled at the channel level when selecting the "Call" type :

Voicemails are also properly handled and can be sent over a "Voicemail" channel: 

Reviewed interactions

Supervisors can now indicate when they've reviewed an interaction, whether they changed any evaluations or not. This will be used to calibrate our model based on the ground-truths provided :

Quality of life features

• It is now possible to delete interactions
• The dashboard creator is now displayed
• It is now possible to export an agent or team performance report

Bugfixes

• The incomplete analyses have been removed from statistics
• The “In production” checkbox has been removed from grid creation

A new release is deployed for Agent Assist (Front-End) on October 9, 2025 (5:30 PM UTC).

Release component : ai-cdn-agent-assist-1.1.0

Release scope : Agent Assist (Front-End)

Release date : October 9, 2025 (5:45 PM UTC)

Release platform(s) : Production

Release content :

• FEATURE : [UI Persistence] Layout preferences and notification panels now persist across page refreshes
• FEATURE : [Tab State] Active tab selection persists across page refreshes
• FIX : [Language Selection] Interface language now prioritizes configuration settings over locale prop
• FIX : [Transcript Display] Harmonized transcript formatting between display and copy-to-clipboard

A new release is deployed for Agent Assist (Back-End) on October 9, 2025 (5:30 PM UTC).

Release component : ai-agent-assist-2.1

Release scope : Agent Assist (Back-End)

Release date : October 9, 2025 (5:30 PM UTC)

Release platform(s) : Production

Release content :

• FEATURE : Added support to headless Agent Assist sessions (Agent Assist features without UI)
• FEATURE : Added improvement for the handling of end of session
• FIX : Fixed regression relative to webhook callback action when triggered after the call
• TECHNICAL : Added callback to properly close all tasks at the end and clean up the state
• TECHNICAL : Added debug to identify potential issues linked to async running in loop (CPU usage)
• TECHNICAL : Migrated the docker image storage to Nebius repo

Team performance dashboard

It is now possible to track team performance in a dedicated view. The displayed data is essentially the same as on the agent performance page, and includes: 

• Agent number
• Number of calls processed on the period
• Average Conformity
• % of calls above 80% conformity
• % of calls equal to 0% conformity
• Average conversation duration
• Best and worst criteria on the selected evaluation
• Latest evaluations
• The radars that we used to have for agents

A search bar is now available on the team and agent performance pages to filter items in the list

Customer platform duplication

It is now possible for a super admin to duplicate a customer platform in order to onboard new clients more efficiently based on a template platform or another customer platform. 

The duplicated data includes:

• Groups of criteria and criteria 
• Groups of evaluation grids and evaluation grids
• Deployments
• Lexical rules
• Tags
• Actions
• Brands
• Channels
• User groups

Note that dashboards, users and interactions are not duplicated.

Quality of life features

• It is now possible to leave a comment on an evaluation
• It is now possible to delete a grid section
• Criteria and evaluation grids are now organized into custom groups
• The user groups associated with a deployment are now displayed in the deployment list
• The “partially compliant” symbol has been replaced with a tilde (~) to avoid ambiguity
• The error message in the option title input field (in criteria) has been improved to clearly indicate forbidden characters

Bugfixes

• Fixed an issue where it would be impossible to select a grid in the interaction list filters
• Fixed an issue where tag names wouldn’t be displayed for people using the QM in another language 
• Fixed an issue where options prompt in a criteria wouldn’t be displayed in the view only mode

Quality of life features

• It is now possible to click on criteria titles inside of grid configurations to access criteria definition

• Added some guardrails to prevent accidentally saving a wrongly formed criterion (special characters etc)
• Gave the ability to modify grid title and groups after being put in production.

Bugfixes

• Fixed an issue where saving a criterion would result in an error if the parameters were changed too fast
• Fixed an issue where the modification of a criterion did not update the "last updated" date.
• Fixed an issue where it would be impossible to add a section to a grid not in production

A new release is deployed for AI App Frontend on July 3, 2025 (2:00 PM UTC).

Release name : ai-app-front-2025.7.1

Release scope : AI App Frontend

Release date : July 3, 2025 (2:00 PM UTC)

Release platform(s) : Production

Release content :

• UI : Completed UI overhaul (Profile, Developer, Model, Home, Auth)
• Routing : Updated redirect links to Diabolocom website

A new release is deployed for AI Core API on May 22, 2025 (5:30 PM UTC).

Release name : ai-core-api-2025.5.1

Release scope : AI Core API (Web Backend of AI App)

Release date : May 22, 2025 (5:30 PM UTC)

Release platform(s) : Production

Release content :

• FEATURE : AI Accounts are unified if they're coming from the same Engage Account, even if they're using both QM and AI App
• FIX : Calls coming from QM should be now all properly indexed in meilisearch
• CHORE : Rate limit has been increased to 1200 requests per minute if calls are coming from our microservices or Engage
• FEATURE : Added callbacks destinations on Agent Assist Settings
• FIX : Parts of transcript could be cropped before, it shouldn't be the case anymore

A new release is deployed for AI CDN Frontend on April 15, 2025 (9:00 AM UTC).

Release name : ai-cdn-2025.4.2

Release scope : AI CDN Frontend

Release repository : agent-assist-interface

Release date : April 15, 2025 (9:00 AM UTC)

Release platform(s) : Production

Web Components details :

• Component : ai-agent-assist-configurator
• Version : 0.1.0

Release content :

• FEATURE : Integrated Sentry monitoring
• FEATURE : UI overhaul and harmonization
• TEST : Improved test coverage up to 70%

A new release is deployed for AI CDN Frontend on April 15, 2025 (9:00 AM UTC).

Release name : ai-cdn-2025.4.1

Release scope : AI CDN Frontend

Release repository : agent-assist-interface

Release date : April 15, 2025 (9:00 AM UTC)

Release platform(s) : Production

Web Components details :

• Component : ai-agent-assist
• Version : 0.1.2

Release content :

• FEATURE : Integrated Sentry monitoring
• FEATURE : UI overhaul and harmonization
• TEST : Improved test coverage up to 68%