Four AI products for instrument, fashion, retail and power-distribution clients
Each product runs in production with a measured result, from a 25% cut in maintenance cost to a 90% improvement in forecasting accuracy.
Predictive maintenance
The challenge
The client is Austria's largest manufacturer and distributor of innovative analytical instruments. Their instruments help customers monitor the performance of manufacturing equipment, and one of their products, EasyDens, is used by home brewers. They asked us to build an AI product that could predict when maintenance would be needed — delivering major cost savings over routine, time-based preventive maintenance.
The solution
We delivered an AI predictive-maintenance product that moved the client and their customers out of a reactive cycle and into preventive and predictive maintenance, with markedly improved maintenance planning.
Results
Technologies used
Computer-vision body measurement
The challenge
This US-based client runs a size-recommendation and visualization platform for fashion eCommerce. They wanted to extend their product's ability to find the perfect fit for the shoppers of the brands they serve.
The solution
Their computer-vision body-measurement AI tool now helps their brands increase conversion and reduce returns. It makes trying on clothes faster and more accessible, including outside the dressing room, and has helped the client's brands reduce returns and create happier customers.
Results
Technologies used
Multi-channel attribution
The challenge
This India-based eCommerce retailer wanted to identify high-value customers who were most likely to buy on their platform, so they could optimize each customer's journey, make the most of their marketing resources, and grow profit.
The solution
The product first identifies the most important customers, then maps how those customers interact with the brand across different channels. It assigns a value of importance to each channel and, finally, attributes credit for each conversion to the channels that contributed most.
Results
Technologies used
Manpower forecasting
The challenge
This Indian client is a power distribution and support-services company. An unforeseen spike in demand for electrical technicians caused their service level to drop dramatically. They needed to predict manpower requirements for their technicians and linemen 30 days into the future, to raise their service level while keeping costs under control.
The solution
We helped them build a product that forecasts their manpower requirements for the days ahead using deep-learning techniques, giving them the visibility to plan staffing with confidence.
Results
Technologies used
Frequently asked questions
What AI products did Mindrops build for these clients?
Which technologies and models were used?
What results did the AI products deliver?
How does the manpower-forecasting product work?
Could Mindrops build an AI product for us?
Contact us
Tell us the outcome you're chasing and we'll show you how we'd build the AI to get there.