What is our primary use case?
We use it for analyzing data and creating models. We extract information from the database and then see if Darwin can share information with us about what would be nice components for the model. Then we use Darwin to make a model. We clean the data and pass it through to Darwin and Darwin generates a best model.
From Darwin, we get parameters, important features, and predictions. We don't have the entire Darwin solution. We just have the core. We are taking the information about the parameters of the model and then we generate the model again with our own tools. Darwin doesn't give us the actual model to use, just the parameters.
We work with Darwin through a webpage and create models there and do linking analysis of the data. We are also working in the SDK version. We connect with the cloud, through the console.
How has it helped my organization?
When you start to use Darwin, you realize how the parameters improve the model. Darwin uses an algorithm that creates 10 or 15 generations and gives you a really accurate model; a more accurate model than we used to make. When we have a clean dataset, within two to three hours we have a really nice model, one that is better than we could generate in a week. A couple of hours with Darwin is like a week of work for us. That's where it provides the most value: in saving time for us.
What is most valuable?
The most valuable feature is the model-generation. With a nice dataset, Darwin gives you a nice model. That's a really nice feature because, if we're doing that ourselves, it's trial and error; we change the parameters a little and try again. We save time by just giving the dataset to Darwin and letting Darwin generate a model. We find the models it generates are good; better than we can generate.
Another feature that is really nice is that Darwin gives you a first impression of a dataset, even if the dataset is a bit dirty. Darwin can give you information about a particular column or feature where there is a lack of information. As a result, we know to do another round of manual cleaning. That's a helpful feature too.
Darwin has also increased our productivity, maybe by as much as 20 percent.
What needs improvement?
An area where Darwin might be a little weak is its automatic assessment of the quality of datasets. The first results it produces in this area are good, but in our experience, we have found that extra analysis is needed to produce an extra-clean set of data. Where it's good, for example, is if we have, say, a date column with different dates and maybe that data is not so valuable for the model because the difference in the dates is not significant. Darwin will find that kind of thing. But you definitely can't give Darwin a dirty dataset and then generate a really nice model. So we have to do extra analysis of the data. Cleaning the data always consumes a lot of time.
Also, Darwin can generate new data, but I didn't find that to be very valuable. Darwin could improve generating new data and that would be an important improvement.
For how long have I used the solution?
We've been using Darwin for about three months.
What do I think about the stability of the solution?
For us, the stability has been good. But when we were training in Darwin, it was not really clear if a model was complete or was still running. They can increase the stability in that sense. One or two times, it seemed like it was not running. It stopped. We waited a couple of minutes and then it was ready.
How are customer service and technical support?
We haven't needed to use support. We haven't had any big problems with it.
Which solution did I use previously and why did I switch?
We did not have a previous solution. I'm just an engineer. My manager came to me and said, "Now we have Darwin," and we started to use it.
How was the initial setup?
The web-based interface is really easy for setup, but the SDK is not easy, although it's also not really complex. I have a big PDF from them with a lot of information but it's not structured in a convenient way. For a first-time user, it's kind of complex. The documentation is not that clear or easy.
Our deployment of the SDK, the first time, took two or three days, although we didn't work on it the entire day. We tried to move it forward a little bit at a time. The web platform is really easy. That took a couple of hours.
What about the implementation team?
We deployed it ourselves. We have a lot of interns in the group so we developed it on our own.
What's my experience with pricing, setup cost, and licensing?
I believe our cost is $1,000 per month.
What other advice do I have?
My advice is to do extra cleaning of your data. Darwin is good when it has a really nice, clean dataset to generate a model, but you need to work at it to make sure you have that kind of dataset.
On our team there are 25 people but there are just two of us using Darwin, my partner and me. He is a data scientist and I am an artificial intelligence engineer.
We are using Darwin for the development phase, but we aren't using it for production. It's a fast tool for development. Within our group in the company, we develop solutions. We try to analyze the possibilities for doing so. We need the data so we extract it and then generate a model. Once the model is ready we put in an API or the cloud or the web. We can then query the model with new data and create a forecast, but it depends on the solution and on the data in the production phase.
I believe we have a one-year licensing agreement. We are trying Darwin to see how it works and its benefits for us. It's hard to say if we will continue using Darwin. We are trying to determine if Darwin is a high-value tool. We need to use Darwin more. I have only been using it during about 5 or 10 percent of my time.
I would rate Darwin at seven out of 10. Darwin saves us some work, but we also have to do extra work. It doesn't do all the work for you. In the beginning, when we started to see how Darwin works, we thought that maybe, from raw, dirty data, we could generate a model really fast, but that's not true. It's good at doing some parts of the work, but you need to work and to think about the solution to your problem. You need to think about the application to generate data according to your solution. Maybe that's a good thing. If Darwin did everything, perhaps I would not be needed in the company.