Amazon Kendra is a highly accurate and easy to use enterprise search service that’s powered by machine learning. Kendra enables developers to add search capabilities to their applications so their end users can discover information stored within the vast amount of content spread across their company. This includes data from manuals, research reports, FAQs, HR documentation, customer service guides, and is found across various systems such as file systems, web sites, Box, DropBox, Salesforce, SharePoint, relational databases, Amazon S3, and more. When you type a question, the service uses machine learning algorithms to understand the context and return the most relevant results, whether that be a precise answer or an entire document. For example, you can ask a question like "How much is the cash reward on the corporate credit card?” and Kendra will map to the relevant documents and return a specific answer like “2%”. Kendra provides sample code so that you can get started quickly and easily integrate highly accurate search into your new or existing applications.
We use it for search, enterprise search, text content generation, and content writing. These are some very high-level use cases that we are exploring with different services.
There are many valuable features. For example, there are many documents that contain a lot of legal information. So we want to understand whether all the documents have the required complaint-related information or not, and whether they are following the standard policies of documentation. We have multiple documents, so we don't know which document has the sought-after information. Therefore, we want to perform an enterprise search on it. So there are a lot of use cases we are trying to build using these newer technologies, specifically Kendra.
Moreover, Kendra has AI, which has an upper edge, and that is really helpful. It has a nice AI inbuilt, which improves the search part of it.
There are some token limits. We cannot ask questions with more than 30 tokens. Access cannot be more than 200 tokens. And the token is also, like, one point. Then views are very hard limits, and it is difficult to increase them. So that is one part Kendra can improve on.
I have been working with Amazon Kendra for five to six months.
We have good use cases where stability is everything. So it's a stable solution.
So far, we don't have that complexity with scalability. At least I have not come across it.
The only scalability challenge I've noticed is the limitation in the number of tokens. Amazon Kendra can only return 100 search results, and that the maximum length of a query is 30 words.Otherwise, it does a decent job.
The customer service and support are good because it depends upon the support plan. So we have 24/7 telephonic support available. So for us, the support experience is good. So, customer-wise, it depends on that also. With 24/7 support, we ensure that SLAs are met, and we are very proactive in that regard.
If the support are not able to solve an issue, they'll involve someone else from their internal team, and they'll ensure that you get the right answer all the time.
Positive
The installation is particularly an AWS service. So no challenges. Simply, all are very well-versed with AWS. Not a challenge for us. So, it's a good process.
The deployment depends on the use case and can be done in less than a month.
For maintenance, it is in AWS service, it has all integration in place. So not a challenge. So if you are going with a multi-cloud approach, then there can be some challenges. But it has all the required logging information, maintenance, and all those concepts are in place. Therefore, it is relatively easy, but if you talk about a multi-cloud environment, it could have difficulties.
The pricing falls in the medium range. The cost depends on the size of your use case because it has a fixed cost, not a variable. The licensing is on a monthly basis.
There are no extra costs. Only the service pack is required.
I have worked with Google and Azure. For me, all these solutions are equivalent; there's nothing like one is significantly better than the others. Because it depends upon the end customer and their approach.
It all depends on the use case and what the user wants to achieve. There are multiple solutions available, even within AWS. It depends on the specific use case and how the end customer is going to utilize it.
Kendra gives you the flexibility and capability to search intelligently, which is its strength. However, in the broader context, it depends on various factors. If we focus on the search part, I would rate it around eight out of ten. There are certain limitations in Kendra.
