JP discusses his virtual Computer Vision company that spans Germany, Nashville, Islamabad and more, and caters to very large customers in Logistics, Retail, and Healthcare.
Sramana Mitra: Let’s start by introducing our audience to yourself as well as to Darvis.
Jan-Philipp Mohr: I started Darvis about seven years ago. At that time, we were still called Hashplay. We pivoted in 2019. Since then, we concentrated on building the ultimate visibility platform for spaces. We use computer vision to translate the real world into data. We do that in logistics and hospital use cases.
We saw a lot of demand, especially where there’s no data generated right now due to regulations, availability, and other solutions being there and being very inefficient. Today, we are close to a hundred people.
Sramana Mitra: Where are you located?
Jan-Philipp Mohr: We are in Germany and in Nashville. We are in Islamabad and London. We have a couple of people in Greece as well.
Sramana Mitra: What is the genesis of the company? It sounds like it’s a global team.
Jan-Philipp Mohr: We started in San Francisco. I’m based originally out of Hamburg. My co-founder is from Nashville. Our CTO is from New Hampshire.
Sramana Mitra: You’re a virtual company.
Jan-Philipp Mohr: Yes, but we do have proper legal structures in place in every jurisdiction.
Sramana Mitra: Talk to me about use cases specific to AI.
Jan-Philipp Mohr: We turn the real world into data. We do that by extracting that data from imagery. We use CCTV and we have AI algorithms that understand the real world based on classification and segmentation to understand the positions of certain items, whether you’re about to fall out of the bed. We created a four-tier alarm system that alerts nurses. There is a big shortage of nurses across the world. We help them to go where they have to be.
Sramana Mitra: Are hospitals your largest segment?
Jan-Philipp Mohr: No, it’s actually logistics. It’s 50/50 more, but the hospital sales cycles are simply longer. In healthcare, you also have to go through a lot more regulations.
Sramana Mitra: So retail and logistics is a bigger segment for you?
Jan-Philipp Mohr: Yes.
Sramana Mitra: What is the use case there?
Jan-Philipp Mohr: First of all, we track assets. We created a platform that translates whatever is in these big warehouses. One of our largest customers is the US Postal Service. We basically track everything in these warehouses. We are deployed in multiple locations. It’s all about how long certain items are idle and what the contents are of these idle items. We track many different things.
There are other solutions like RFID which are simply not feasible for them. They would need to put tags on a million things. They have a lot of fluctuation so they need something that is very expandable and very optical.
On top of that, we calculate the meta status of things. We calculate fill levels. We read bar codes, all on the fly. Then we put that on a map. We do that same thing with other assets in the hospital. In the hospital space, we call it the auto-pilot for nurses and staff to put visibility where there should be visibility.
Sramana Mitra: How much of this is product and how much service and system integration do you need to deploy a large use case?
Jan-Philipp Mohr: It’s very much product. The training aspect is integrated into the platform. The customers can do that themselves. We go further down the pipeline. You can deploy the models on site. One big differentiator for us is, our stuff is deployed on the edge or in the customer facility. We are not running in the cloud. We can because we control the entire stack, but we saw that the customers we’re dealing with prefer to control their own security and environment.
Sramana Mitra: This is an on-premise deployment?
Jan-Philipp Mohr: Yes.