This is a dev-centred review of the product. We've deployed it on a large post-secondary site with over 10,000 indexed pages. When we chose Airgentic, it offered some of the best options for controlling the agent responses that we had seen in a survey of available options.

Since deploying the software, the feature set for tuning and controlling the responses has grown quickly, often with our direct input and requests. We have all of the tools necessary to continue tuning the responses and many of the administrative settings will be immediately familiar to those who have worked with RAG/AI search systems and chatbots.

If you don't want that level of detail, the core product works out of the box very well and you do not need to be a prompt engineer to get good responses. There are many ways to optimize performance, from curated responses to controlling which content is used for answers. The software meets the user. It can be as easy or as technical as you want it to be and this is fantastic for our team of devs, analytics experts and content owners.

The support level has been excellent and the initial deployment is straight forward and easy to get started.

In the case of bugs and issues, Airgentic was quick to help us determine if it was a platform problem or something of our own doing. In the former case, they were quick to provide detailed analysis and fixes where required.

We have already seen a clear positive signal for ROI, but I'll leave that review for our analytics experts. ;) Review collected by and hosted on G2.com.

We have had some glitches but these are transient and typical to any platform that utilizes the underlying LLM technology and similar to what we've seen prior with other platforms. See the corresponding note about bugs and issues in the positive side.

I really have no 'dislike' specific to this product. Review collected by and hosted on G2.com.