We use Machine Learning and Data Science to get answers from data
We use Machine Learning and video analysis algorithms to gain a high-level understanding from digital images, similar to that of a human.
Detecting and tracking objects, faces and gestures in images
Natural Language Processing
NLP allows our algorithms to understand and infer meaning from human language.
Interpreting and classifying human generated content.
A field of Machine Learning methods which focuses on learning data representation.
Automating recognition of features and patterns.
One of our key interests is combining scientific methods and algorithms in order to retrieve knowledge from disorderly, raw data.
Discovering patterns in large data sets.
ML can help with churn prediction and spotting suspicious activity like: Fraud detection, Money Laundering attempts, biometric spoofing attacks and phishing scams.
Adaptation and filtering
Using statistical algorithms we can allow apps to quickly adapt to new situations and implement context-sensitive, personalized filters.
As more data gets processed by production systems, Data Science can be employed to discover patterns — resulting in better predictions and improving the decision making process.
Many e-commerce and content publishing experiences can be augmented with ML-based recommendation engines specifically tailored for the brand, market and user segment.
Cyclops – AI-powered stalking and tailgating detection app
An application developed for our partner in the intelligence and law-enforcement industry to gather data about suspicious driving. We made real-time car tracking and car identification possible with a custom-built Deep Neural Network.
butik – a new way of shopping
An automated, smart vending machine powered by computer vision. Uses a single facial recognition camera for access control and a secondary camera for inventory tracking. Product detection enabled by a state-of-the-art convolutional neural network with bounding box regression and classification.
FaceSecure – spoofing detection app
A lightweight library based on Apple’s CoreML technology to detect spoofing attacks: attempts to acquire someone's privileges or access rights using photo or video of the authorized person’s face.
A conference & festival, fully-automated smart beer dispenser using face recognition technology for ordering, payment and checkout. Developed in partnership with Wirecard for Digital Life Design Berlin 2017 conference.
Toxic comment classification
We regularly take part in challenges to stay on top of the industry and in fields relevant to our expertise. We participated in “Identify and classify toxic online comments” challenge aimed at improving the auto-moderation features of online communities.
Quasar is dedicated analysis software that helps to predict energy demand in order to optimize power plant utilisation. YND helped to build a stochastic model for optimizing trading decisions for hydro storage systems that uses approximations of past trends and historical data.
Read more about Machine Learning in our case studies
Anti-Spoofing Mechanisms in Face Recognition Based on DNN
Imagine being able to unlock your phone just by using your face, no fingerprint scanning or touching required. It would just work, automagically and flawlessly without any user intervention. Wouldn’t that be great?
Car Detection & Recognition Using DNN Networks
License plate detection is a common use case which has been solved several times in the past, but felt that we could provide something better than the current status quo.
A New Way of Shopping
When we think of shopping, we often see it as an annoyance with lots of waiting involved. What if you could just walk in, take what you need and simply walk away?
Started From The Bottom, Now There’s Beer
Wouldn’t it be cool to just be able to walk over to a bar and not have to fight for attention to get your drink? With SmartBar, we’re quenching your thirst the smart way.
Industries we focus on
Whether you're a multi-million dollar travel company looking to innovate or a small FinTech startup launching an app, we don't discriminate. We work with businesses of different types and sizes, and we excel in these industries:
Trust is the source code for great products
We don't just build products. We also invest in building long lasting partnerships, and as a result our clients often work with us for several years. Our work goes beyond building apps, shipping products or developing hardware. Our goal is to work as an extension of our clients' team as their trusted partner of choice.