With a goal to grow and improve its machine learning (ML) capabilities while keeping the highest security and reliability practices, New10 was lacking one unified standard for its entire ML process. It had been using Amazon Redshift to store its data and several other tools for processing and analysis, but Eraneos identified what the next steps should be: to help them in becoming more data-driven. We knew this could be done by further leveraging the cloud technologies of AWS, and more specifically, AWS Sagemaker.
The first port of call was a proof of concept that demonstrated the capabilities of AWS Sagemaker and what it could mean for New10. On the engineering side, we built the necessary infrastructure and security regulations in line with the company’s requirements. Sagemaker studio was set up on the data science side, which allowed us to build a process that would serve as a guidebook when training new machine learning models.
Through our company training, we also set up an AWS Sagemaker course for New10’s data team, which let them get acquainted with the environment while teaching them how to use the tool to its full potential.
New10 is an initiative of ABN AMRO that combines the flexibility of a startup with the reliability and experience of a bank. Its goal is to help entrepreneurs and small businesses accelerate their growth by offering them financing and financial resources. Starting from a small team in 2016, it is now helping thousands of entrepreneurs and their companies. New10 can arrange everything online with financing ranging from €5,000 to €1 million and attractive annual interest rates.
Using AWS Sagemaker alongside Eraneos’ support, New10 is now able to build and train Advanced AI and ML models in one central location. The end-to-end machine learning solution also means the company’s data team can store, process, and prepare data for analysis without needing programming skills. This allows it to quickly and efficiently test different business cases and put them into production. In terms of data science, the team can work faster and more efficiently on data transformations and visualizations, as well as spend less time on data quality and cleansing.