To reduce false alarms significantly is the task of the Berlin-based startup promiseQ. Their solution is to use the strenths of Artificial Intelligence (AI), but to complement them with the skills of humans. The two founders Tolga Ermis and Elias Kardel told us how exactly this is supposed to work and how the idea came up. Read more in the March issue of MTH Unboxing.
MTH Accelerator: What did you do before promiseQ?
promiseQ: Tolga studied industrial engineering, focusing on product development. Elias studied electrical engineering, focusing on new technologies and algorithms.
We met at HELLA Aglaia, a Volkswagen subsidiary that develops computer vision for cars. Our team was working to integrate artificial intelligence into a range of sensors and interfaces used in camera-based driving functions such as brake assist, lane keeping, and multi-touch cameras. Working in the final phase of the development pipeline, we were able to meticulously examine (and admire) all the abilities of artificial intelligence. However, we also became aware of the numerous errors and safety gaps that can arise from a lack of monitoring.
As part of a pilot project, we investigated the benefits of human involvement in cases when the analyses by artificial intelligence were unsuccessful (for example, recognizing an oncoming vehicle or a road sign). The project was eventually terminated by the leaders at HELLA, but we had discovered something great during our time working on it.
Subsequently, Elias left for Taiwan, and Tolga took a job at IAV, another Volkswagen subsidiary. There, Tolga conducted test drives to evaluate front camera sensing. The benefits of human involvement in such processes were unmistakable, so we started discussing the issue again. We talked about how to apply crowdworking to problems in areas such as quality control and healthcare.
MTH Accelerator: What do you do in your free time?
Tolga: I do a lot of sports to rebalance myself (HIIT, jogging, Thai boxing, BJJ––I have a black belt in Taekwondo). I also like to read and watch movies when I have the time.
Elias: I love Asia. I lived in Asia for a year and a half, and I practice Chinese sword fighting. After a long day of software development, I also like to work with my hands in my (hobby!) wood workshop
MTH Accelerator: What is promiseQ?
promiseQ: promiseQ is a tech startup founded in Berlin. We have developed a cloud-based platform that verifies and filters false alarms in real time through AI and an on-demand human-in-loop service. The promiseQ hybrid engine complements AI, which continuously learns with human-in-loop verification. This method allows us to take advantage of both worlds: machines and humans.
The increasing popularity of crowdwork and modern cloud technologies allows us to scale as needed. In addition, by using the human-in-loop approach, we can gain a significant competitive edge for various scenarios outside the scope of traditional security applications. The promiseQ hybrid engine can be used to monitor business protocols and processes (i.e. correct behavior in the correct order), detect hazardous situations (i.e. accidents, fights, theft), monitor inventory, detect contamination in the food and cosmetics industry, detect wildlife on farms, protect artifacts in museums, and much more. The possibilities of promiseQ’s hybrid approach are endless!
MTH Accelerator: Why the MTH Accelerator?
promiseQ: Through mentoring and coaching, MTH Accelerator has given us the opportunity to improve our activities in business development and sales. We have also been able to expand our network with MTH Accelerator’s community of supporters and alumni. This in turn supports our fundraising activities.
MTH Accelerator: Why promiseQ?
promiseQ: Right now, crowdworking as a way to provide security is still a novelty in our rather conservative industry. In a world of surveillance centers and automated solutions, crowdworking to protect an infrastructure is a completely new idea. But the numbers speak for themselves: in certified pilot projects, promiseQ has achieved a 99.5% success rate in reducing false alarms and has done so at a low cost. The suggestion of decentralized security monitoring may sound far-fetched––just like the idea did at one time that cars could move faster than horses––but the success rates and cost savings show us that it is anything but.
MTH Accelerator: What is your mission?
promiseQ: Our mission is to reduce or eliminate the number of incoming false alarms so that our clients can focus on their core business of creating security and adding value for their customers.
The MTH Accelerator portfolio is growing and growing. But what exactly do the teams do and, above all, what did they do before they became entrepreneurs? In MTH Unboxing, our new interview series, we shed some light on the subject and introduce the teams in more detail. MTH Accelerator Babelsberg is the program for startups and founding teams from the metropolitan region who are looking to explore technologies and digital business models and take their companies to the next level.
Learn more about the program, our latest portfolio startups and events in our newsletter.
