According to Gartner, over 50% of machine learning models do not make it to production. One of the main factors for this is that once a model is put in production it needs constant monitoring, retraining, and re-deployment because of changing data or changing business needs. This is the case for the manufacturing industry where sensors for multiple machines can be used to build predictive models for several measures such as energy consumption or for maintenance prediction. The sheer quantity of sensors and machines makes this task too cumbersome for any company.

Thinking of car commercials - safety, technology, and luxury - may cross someone‘s mind. The ads aim to create feelings such as excitement, fulfillment, and thrill. Transferring these expectations and feelings into the real world is a challenge. Among existing sports car owners, only a small fraction engage in performance driving. Unawareness of performance-driving opportunities is one reason, in combination with the doubts about accessibility. Overcoming these obstacles is the key to reducing the gap between the anticipation and actual experience of sports cars.

Schaeffler wanted to motivate its employees to come to the office more frequently, to foster collaboration and team culture. To this end, deskbird was introduced as a week planning and desk booking solution to make the workplace experience more predictable for Schaeffler‘s employees and remove uncertainty about who is in the office and whether there is enough space available on any given day.

The main challenge was leveraging personalized AI videos in the content creation process at Bosch to enable video production at scale. The goal was to reduce time and cost associated with video production, while increasing audience engagement by tailoring the content in multiple languages with personalized digital avatars.