DataRobot

If you're starting to build out your Data Science + ML/AI team this is the platform which will jump start you and get you from 0 to 1 much faster compared to doing things natively within the hyperscaler eco-system.

For very complicated use cases or where you need hundreds of models at a micro geo level the licensing may become cost prohibitive.

Well suited to regression and forecasting.

DataRobot is great for data scientists to allow them to work faster. But the simplicity in creating models could hurt a company if someone doesn't understand the model.

Data Robot is a powerful tool for greatly reducing the time required to build powerful and accurate machine learning models. It then allows you to utilize these items.

It is probably most appropriate for organisations looking to get into data science and incorporate Machine learning and AI into their decision making. Having dedicated resources that can be upskilled is perfect, as the expertise and software provided allows for a big jump from willing to able.

For the to work effectively, organisations should really consider dedicating at least one resource to the ML and AI projects, and understsand that not every project will yield fruit. A lot of this is innovation and experimentation, so relying on data Robots insights in make or break situations is not recommended. You also need to manage expectations well as the data you have may simply not allow for a powerful model.

Finally, the organisation must be open to change, this has to exist in tandem with the above. If the organisation's key stakeholders don't want to change, all the insights in the world won't help. So a willingness and ability to change effectively is required to maximize ROI.

DataRobot is well suited for generating models, such as the Bscore model, which is very important to the banking and financial industry. DataRobot provides insights and recommendations as to what are the relevant parameters for predicting the score of every customer based on their behavior. DataRobot also cut the time on making a model for forecasting. I use DataRobot for forecasting the probability of default using the ODR and linking it with economic indicators.

Most problems I've encountered in my career can be framed as supervised machine learning problems, and adequately solved with a fairly common workflow and popular ML model families such as xgboost and lightgbm. DataRobot is one of the most low barriers to entry but still complete solutions I have encountered. I appreciate the autopilot system and found that it is an excellent starting point for a project and in some instances running it in comprehensive mode is sufficient to arrive a deployment-ready model that can fit into a BAU. Even entry-level data scientists would be able to hit the ground running with DataRobot and produce a lot of value for their organisation. The only aspect that might limit utility is data preparation, especially domain-specific requirements and sensitive data that cannot be uploaded in a raw form to a cloud hosted service outside of Australia due to privacy concerns.

It is best suited to be used by Data Analytics team, especially in data modelling. It saved a lot of time since there were already recommended model to choose from.

Predictive Modeling. Using Datarobot, I was able to build accurate predictive models quickly. It is also very useful in shortlisting risk factors, it provides Feature associations to include only the most relevant features in final model to reduce complexity and improve interpretability.

Well suited for companies with mature data sets to enable quick project creation. The product enables discovery of new features based on the impact to overall model. DataRobot is excellent at reducing time to market for modeling projects.

In my opinion, DataRobot is well suited when you have completed preparing the data into a relational format. The model screening and estimated ideas on the ranges of accuracy or errors. For beginners or intermediate users, DataRobot University is also excellent. On the other side, I found it less appropriate for any data cleaning or EDA task.

It's very well suited, in my opinion, for small teams or those cases well your models are necessary to support internal business decisions. Probably is not the best tool if the product that you are selling really is an ML model; in that case, you can use it as a baseline, and then you can try to beat Datarobot results.

The large amount of household specific data with huge feature space is great for DataRobot to run. The result visualization is also easy to follow. However it would be better if there are more NN models could be implemented if possible.

We have a variety of complex use cases in manufacturing & supply chain which require creative use of Regression and Time Series with highly engineered features. DataRobot has done training & deploying models a breeze! The Retraining functionality for deployed models is also very useful.

DataRobot saves the user an enormous amount of time by automatically implementing best practices related to feature management, partitioning data and evaluating models. DataRobot also provides a "learn on the job" experience in that users with limited knowledge and experience can immediately get useful results and then learn later about the technical details of modeling and handling data.

Time series data. Honestly, there's a lot of information that gets collected over time. Using DR, it becomes easier to track what's going on and get to the point where we're predicting what's going to happen. While DR is great at the big picture stuff, a lot of times (for quick/easy answers), it's better to just do the quick math yourself.

DataRobot is a Powerful tool for teams that are beginning in datascience world where learning curve and speed are critical in order to generate value within any organization.

Great for straightforward predictive analytic work, such as the quintessential business cases (e.g. churn prediction). Not so good for unsupervised stuff, such as clustering, and it is pretty hard to use to predict recommendations.

If one takes the time to learn the platform, the platform can be very useful for making predictions. It's not perfect, and you do need your real-world insight to determine if their predictions have the potential to be better than an internal system or a rudimentary system or are wildly off, but for our business, just a slight improvement can mean thousands of dollars in both revenue and thousands of dollars of savings from not purchasing items we may have otherwise purchased.

DataRobot is a very robust and easy-to-use tool. It can be used for practically anything. It works best in cases where you have powerful predictors that are not highly correlated with each other, and you are looking to predict a single outcome variable. If you are trying to predict multiple variables at the same time, you will have to create separate models for each one. While it does provide information about the effectiveness of each predictor, it will not give you a detailed analysis of the interactions between your variables. However, it will let you know if some variables are redundant so they can be removed.

It's appropriate for speeding up the work of your experienced data scientists. If they spend more than 15% of their time building and tweaking models, DataRobot will cut that down significantly. Caveat emptor: while the DataRobot marketing materials promise to turn any analyst into a data scientist, this is far from the truth. If your potential users do not already understand how machine learning models work, and have not built some models on their own, then they will make mistakes that DataRobot will not correct because it assumes you know what you're doing. Interpreting the results and iterating on models is easy for a trained data scientist but would be baffling for a typical financial analyst.

DataRobot is tailored to efficiently automate algorithm generation and model testing based on industry-specific best practices. It is user-friendly, well-designed, and capable enough to develop world-class classification predictive models with a reliable monitoring interface.

It is well suited to carry out forecasts and feed from massive amounts of data to simplify decision-making in a seamless, comprehensive, and applicable manner.

Great time saver for quick exploratory analysis or testing ideas and possible solutions. Not so good for working with very sensitive data or when additional data anonymization is required.

From a credit risk perspective, I think DataRobot is very useful to work on scenario assessments. For example, to identify the worst possible scenario in a default assessment.

It is well-suited when you can finally be able to work "easily" when you know how it works. It is not suited when you start to use the tool you need to contract a consultancy, but you have to know how to use it. Also, they have courses you can use, but you have to dedicate the time to study and learn.

Datarobot is a super powerful and useful tool for all data science teams, especially for those teams which are not specialized but where all team members take care of data products from A to Z ( from data foundation to prescriptive analytics) as the tool is democratizing AI. It is very useful as well to gain time, you do not need to spend hours and hours coding as the platform does it for you already. During the training they are providing, you can experience on your own, that regardless of all the manipulation you can do to the models on your own, you will never beat the machine