AutoQA GenAI Behaviours

With the advent of LLMs and advanced AI - it was only a matter of time till we provided you with GenAI powered AutoQA behaviours. These behaviours are potentially the biggest enhancement to AutoQA for a multitude of reasons. Such as - 

• Context - These models are more advanced and need less context and work properly over a much larger set of data
• Transparency - These models are smarter and are able to provide more transparency and reasoning
• Insights - These models are able to output qualitative and actionable insights

That, along with certain limitations of AutoQA when using our in-house models, listed below, are where and how these new behaviours add significant value - 

• Output - Currently the output is only a grade - Positive/Negative. No narrative around reasons or improvement.
• Explainability/highlighting - Currently only available for some models, not all. 
• Generalistaion - From time to time, we come across some cases/language etc, that isn't aligned with the model and the model needs tweaks / training. 

As such, all of our new GenAI based models to come will come with two additional features - 

1. Reason Statements - Irrespective of the grade assigned, AutoQA will also output a generated statement that explains the reason behind the grade provided.
   
   
2. Coaching Feedback - This is a generated statement that is only provided for Negative grades, that lets the agent know what could have or should have been done to get a positive grade instead. 

Rollout - This is being incrementally worked on and enhanced, ie we're upgrading ~4 behaviours per quarter, till our entire AutoQA scorecard is GenAI powered and is being currently released to beta customers only. We're approaching this from the point-of-view of what is the most requested and complex. If you're keen on trying this out (during the beta) or if you'd like us to prioritise one behaviour over another, please reach out to us via directly or email.