Below you will find answers to our most frequently asked questions. If you cannot find the answer you are looking for, please contact us using the FormOpen external link
If you are interested in our product, send us an email at [email protected] or via FormOpen external link and we will provide you with a token to carry out experimental tests. After testing, if you wish, we can schedule a video call to learn more about you and your company, and explore how we can further assist your business.
This is very important! The minimum acceptable is between 100 and 200 words. To have greater accuracy of the model, the ideal is for the text to have a variability of words, more than 500 words is a stronger analysis.
All our services are built on a serverless architecture. It’s important to note that during the initial analysis, the response time might take slightly longer, up to 20 seconds, as the service ‘warms up.’ After that, the response time typically averages around 5 seconds per prediction when using GPU acceleration. If you require even faster response times, we can create an environment tailored to your specific needs.
Our model underwent primary training in English and Brazilian Portuguese, alongside the development of an extended multilingual version. We emphasize refraining from the use of translation services, as they may impact model performance. If you need a model in your particular language, please reach out to us via the contact FormOpen external link, and we can explore customized training solutions. Our service adeptly identifies the language of the customer’s text and routes it to the relevant template, be it in English, Brazilian Portuguese, or our multilingual version.
The Power Skills service extracts the core competencies of the future professional. We treat Power Skills as another product.
Do you create personalized result analysis for my company?
Certainly! We can create customized analyses and reports for your company based on the generated results, including presentations for your team members involved.
The scores for all dimensions and sub-facets are the raw-data of the model, we simply percentile them to facilitate understanding, but we do not do any type of processing of the raw data or apply population re-calculation.
Will it be adding other theories to the model besides the Big-Five?
Yes! And we are working on it, but we prefer to treat it as another product.