In 2015, MoBagel officially established its new headquarter in the Silicon Valley, and was accepted into the renowned startup accelerator 500 Startups' Batch 13. MoBagel showcased its IoT analytics solution that uses real-time reporting and predictive analytics to help hardware manufacturers save millions in R&D and operational costs per year was subsequently named VentureBeat's '8 of our favorite startups from 500 Startups' 13th batch'.
Since graduating from 500 Startups, the MoBagel team recruited top talents from Stanford, UC Berkeley, Oxford University and National Taiwan University to further the technical development and business operations.
Through Softbank's Innovation Program and Nokia's Open Innovation Challenge, MoBagel worked on multiple data science projects with different enterprises to enable prediction across industries such as IoT, retail, manufacturing, and banking.
MoBagel learned from its own experience building predictive solutions for enterprises that a lot of the data science work could be automated and that many companies have trouble building their data science solutions due to high costs associated with it.
To solve this problem, MoBagel since developed Decanter AI, a fully automated machine learning (AutoML) engine, to build predictive solutions that directly integrate with clients' systems to convert data into actionable insights that promote business growth.
MoBagel's Decanter is made for businesses that want to utilize data science methods to make better data-driven decisions and extract deeper insights from their valuable data. This usually requires hiring a full-fledged data science team, but MoBagel significantly lowers the barriers by making its data science tool simple enough for anyone to use.
The Decanter General Purpose (GP) Module is designed to automatically build models from raw data without a predetermined scenario. It has the flexibility to be applied to any given situation and ideal for generating predictive insights on-the-go.
The Decanter Employee Turnover Module helps enterprises predict the performances of prospective employees and the likelihood of employee turnover based on candidates' resume, test performances, behavior assessments, and other assessments.
The Decanter Campaign Forecasting Module uses historical campaign and sales data to predict sales performance of future campaigns and optimizes controllable factors to maximize the performance of each campaign.
The Decanter Upselling Prediction Module uses existing customer data to identify potential upselling opportunities for specific promotions, and at the same time also identifies groups of customers that are likely to be lost in the near future. This allows businesses to capture opportunities that can generate more revenue and prevent lost customer revenue.
The Decanter Predictive Maintenance Module is designed for businesses that rely heavily on the performances of machines and equipment. Decanter AI digs through machine logs, error logs, and other machine features to determine underlying error causes and predict error occurrences that aim to achieve zero machine downtime.