Sagitto helps its customers to take advantage of our many years of experience in applied machine learning. Our speciality is building machine learning models for use with spectrometers - from high-end benchtop instruments all the way down to tiny chip-scale spectral sensors. Regardless of the Hardware Option or Account Option chosen, Sagitto clients can subscribe to a service plan in which Sagitto's data science team provides data mining, data cleaning, data modelling, and system customisation specific to your particular application.
Your Data, Your ModelsYou own any predictive models that we build using your data. That means that you determine who gets to use the models - usually our customers wish to retain their proprietary models just for their own use, and we completely support that approach. However if we have customers with similar applications, then we are willing to look for ways that data from multiple customers can be combined (with their permission) for their common good. |
Leveraging Existing DataWe have a lot of data already available - including spectroscopy datasets we at Sagitto have collected ourselves, as well as those collected by our users. If some of this data is also applicable to your problem domain (and if you have permission to use it, in the case of our customers' data), we may be able to use it to power-up your machine learning models - allowing you to reach a higher level of accuracy, or to reach your target accuracy more quickly. If you're looking to kick-start your spectroscopy application, or looking to go further with your current data - or if you're an existing customer interested in an additional way to monetise your datasets - please contact us. |
Managed ServicesMachine learning models are usually not static - the word 'learning' is there for a reason. Sagitto's annual Data Science As A Service subscription not only includes helping you design and implement a suite of predictive models best suited to your problem domain - we also take care of their on-going management. This includes arranging cloud hosting in Microsoft Azure, on-going performance monitoring of model accuracy, and regular retraining to ensure that your models develop and grow with your business. |