Excellent work. Please continue to bring regular videos on gcp agent builder

Love your videos. Keep on teaching such knowledge.

@techwithsach

4 ай бұрын

Thanks, will do! 🙂

Hello! In my case I have a sales table, if I want to ask the assistant how many sales I have in x month or the top 5 best sellers... although the information is in the table, it is better to create webhooks and not import the table from the data warehouse? tanks

My data source is on an IBM Iseries. What tool can I use to build a chatbot to query that data?

Thanks for sharing! This is very helpful Question: I've set up a chat application that utilizes a bigQuery table containing sales data for products and SKUs. However, the chat app isn't responding to questions using the data from the table ( for example: how much we sold last year for a specific sku or product etc..). What do you think the issue here?

@techwithsach

4 ай бұрын

Hi, If you are using Data stores, please see the options that it can be used for. This is from the official documentation. I believe what you are trying to achieve falls in the category what Datastores are not the best option to be used. You can achieve it as a API call or webhook where you are doing this as part of your business logic. If you plan on using Datastores you might need to convert that into a more natural language FAQ kind of a structure to make it work. Hope this helps. The following scenarios are recommended for data store handlers: - Questions that can be answered by your organization's documents or website. - FAQs that do not require database lookups. - The following scenarios are not recommended for data store handlers: Content that does not have answers to desired questions. - Questions that require database lookups or server requests. - Scenarios that require data redaction. - Scenarios that require deterministic agent responses.

@shaikbalajibabu4408

2 ай бұрын

Hey @MoshtavaSaleh were you able to fix the issue? Please respond

It appears the option to select BigQuery as a data source is gone. Now, only web, cloud, and API options remain. :(

Thank you your video is helpful but this vertex interaface system is full of gremlins, its a really ackward process flow and its filled with delays and peculiarities from the datastore. I often just get stuck and cant find a way out of the interfaces so i delete and start a new project. After playing with it for a few days I abandoned it. Personally i find it unworkable.

@techwithsach

4 ай бұрын

Hi, I have faced a similar situation when you try to link Bigquery tables which are not structured properly with natural language text. I have seen success when the table contents have more of a question answer kind of resource especially when you choose Bigquery as a source like I showed in the video. I am sure there might be more changes to this in the future. Also, you can always check if your datastore is properly imported and indexed before you try the app. In addition to this, the Datastore agents should not be used for the following. The following is from the official documentation. Hope this helps. cloud.google.com/dialogflow/vertex/docs/concept/data-store-agent The following scenarios are not recommended for data store handlers: Content that does not have answers to desired questions. Questions that require database lookups or server requests. Scenarios that require data redaction. Scenarios that require deterministic agent responses.

@davidrocha3357

2 ай бұрын

Have you find a solution for this ideia ? I have the same scenario as well and I need a solution for that.

Cut the intro down please. Just some feedback, it is way too long.

Thanks 🎉❤ for making this video. I regularly follow your videos on agent builder apps. I have a couple of questions: * How can I connect a database to a chatbot, specifically if I have a student table in BigQuery? * What are webhooks, and how can I use them to access this data? Could you make a video with an example using a student table on these two topics? It would be very useful for us. Thank you.❤🎉

@manikantareddy3406

29 күн бұрын

Please reply