IBM Watson Studio Reviews

Our primary use for Watson Studio was to analyze the trans-video prints about the operation so they can get their own insight into how the public thinks about their corporation. Also, I analyzed resumes so the AI can generate the proper information for the resume.

The most important thing is that it's a multi-faceted solution. It's a kind of specialist, not a generalist. It can produce very specific information for the customer. It's totally different from Google or any search engine that produces generic information. It's specialty is that it's all on video. 

It's professional.

AI is an important part of Watson Studio.

The product is already really great but for most researchers or a person like me, there are few templates to try something new. So we're limited.

Also, I think to make the most of Watson Studio the user interface needs to be improved. For the developer or the engineer, it could be very easy to use but to a person or a text analyzer, they could have some problems using it. So a better user interface could be very helpful

I used it briefly.

In terms of the stability, it's technology is like a black box. We never know how it works, how the AI engine works. We find out how but to do the work we need to try lots of experiences.

The scalability was great. It worked. It works fine as long as you can train it in the right direction.

I would rate IBM's technical support as average.

The initial setup was straightforward. The deployment took maybe three months to six months because we tested it and tested it, over and over again.

We have our own team that implemented it.

I would rate it around eight or nine out of ten because sometimes I experience some unknown error.

I'm looking at data analytics for division making. In the working environment, I'm looking at how we can use the system to give insights based on company data relating to employees. 

The most valuable feature is the system's ability to take a look at data, segment it and then use that data very differently. What I'm referring to is machine learning which can be applicable for other data sets without having to carry out the process all over again. It's a very powerful feature. In addition, the ability of the product to firstly have the solution as a cloud solution and second, the ease of use of the application. The importation of data and then shipping it for analysis is a very easy and small process.

I'm not yet familiar enough with the product to talk about improvements but one aspect that comes to mind from what I've seen is that sometimes it's easy to get lost given the number of images the solution opens up when you click on the mouse and the amount of different tabs. It might be easy for someone to lose their way around the system. But no real complaints, so far so good. 

Previously we had what you call a multiple document interface so that whatever you do, you're still within the same container. You could look at the same kind of environment where you don't have to jump between different windows. You can split a window if you need to get to some other bit of information in the system. You leave your window, and then reference that information on the same window as opposed to jumping to another window. Otherwise, you end up with too many windows and they get cluttered. It becomes hard to navigate around.

I've been using the solution for the last two months. 

For the couple of months I've been using the system it's been stable. I have the metrics and I have the performance parameters to show this, and I have not had any issues warranting assistance.

I haven't had any experience with technical support. 

The initial setup is very straightforward. I'm impressed. I think as soon as you get logged into the system, everything works, this is especially true with a cloud-based system. But we haven't gone commercial yet, so I'm still using the environment that has been set up by IBM itself. There may be some issues that I haven't experienced yet. 

For now, we're using the light edition which is free but I'm sure once we decide to go commercial with it, we will pay some license for it.

The general advice I would give is that people should think twice about the kind of results that they are seeking. This is general for any system, not only IBM products, but they should be aware of the resources that they're sitting on that is data.

Based on the brief experience I've had with the system, it works for me. It has given me the kind of results that I was looking for. It is stable. I want to believe that the support is equally good even though I've not had any experience as far as that is concerned, but I want to believe that based on the other parameters, the support will follow suit. 

I would rate this product a nine out of 10. 

The primary use case is an algorithm for our radar sensor for autonomous driving. We do use it in a cloud environment for our testing.

When there is more interference, the speed and efficiency of the algorithm become more important. This is where we have high hopes that this technique will help us. Not only in the current interference algorithm that we are working on, but in the future with things like object classification.

It is a stable, reliable product. In our experience, everything is fine.

The product should meet our stability needs going forward.

Technical support is great. We have had weekly teleconferences with the technical people at IBM, and they have been fantastic.

The initial setup was straightforward.

We started looking at IBM with Urban US. That is where we made the connection.

We did look at some open source products, but nothing in particular, as IBM stuck out.

Most important criteria when selecting a vendor: 

• Reputation 
• Being a start up, we need to have good technical support. There are big players that would not even talk with start ups.
• Two-way communication is really important, especially for something as new as this, which is why we have weekly meetings. It is helpful on both sides.

We use Watson Studio with an application we developed for restaurants on the Facebook Messenger platform using Chatbot. It contains many features like making orders and table reservations.

Every feature is good in Watson Studio.

Improving it just like the features in Google Dialogflow and Small Talk. If these features were also in IBM Watson Studio that would work for us.

I have been using IBM Watson Studio around one year.

In terms of stability - it's stable. As compared to Google Dialogflow it's very stable.

The scalability of IBM Watson Studio is great.

Currently, we have around 100+ clients. There around 1000 users using our platform and starting in the new year we have plans to increase our clients.

We now have a six to seven member team maintaining the program. 

In our experience, compared to Dialogflow, they have great support. IBM has great support, so we are certified with IBM also.

We used Google Dialogflow but the setup and running were very complicated. So that's why we moved on to IBM Watson Studio.

The initial set up was very easy, especially as compared to Google Dialogflow.

The deployment does not take much time, around one and a half days. The training takes more time because I have lots of use-cases. It will also take time to train our bot.

I evaluated Google Dialogflow and Amazon. In the end we chose IBM Watson Studio because the training and features in Dialogflow are very complicated.

IBM Watson Studio has great features but the decision making in their decision making feature is less good than other options. If IBM Watson would improve their decision based system, then definitely IBM Watson could perform the role of an AI engine.

I primarily used the solution as an API. The solution was a deployed model and I was just installing the API, sending some data and returning some data from REST API. It was very easy to use.

The computer version was very helpful. It returns approximately five clusters of the projected features.

The solution is very easy to use.

When I'm exploring the data and it takes me to another page, when I want to return to my workflow, it returns me to the first step. I'm not sure if this is typical but it was my experience. It would be great to make it easier to explore the data and complete the workflow using the same method.

More features in data virtualization would be helpful. The solution could use an interactive dashboard that could make exploration easier.

I don't think I did the steps to see if it's mature enough or not and considered stable. You have to use it in the product to see how it performs. In my case, it was just a demo that I used. I was just testing the tool, so I can't say if it's mature or not.

I installed the solution in a stand-alone server. I didn't test for scale virtually or to see if there was any failure so I can't speak to its scalability.

I haven't used technical support.

The initial setup was easy. You just need to follow the demonstration.

We didn't need outside assistance. I have experience in installation and also in Linux and Linux big data, so it was very easy for me to install such tools.

We're an IBM partner.

I'd recommend the solution. I'd rate it eight out of ten.