Technology we build.

At SynerScope we built our products from the ground up. This allows us to design and test the required workflows for unstructured data, making our software suitable for even the most complex data environments.

All that unstructured data knowledge has shaped the way we build our software.

IXIVAULT

Ixivault helps sort, label and categorize dark data, the unstructured & structured data of which you no longer know where it is or what it means. For this we use the flexible infrastructure of the Microsoft Azure cloud, which saves you a lot of time in really getting to know your company data.

Ixivault visualizes linked data not only from traditional sources such as spreadsheets and databases, but also includes data such as images, including video, IoT data and documents. Ixivault helps users to quickly sort, categorize, label and clean large data sets, including text or images. This allows you to make confident decisions about whether to retain and/or discard data.

IXIMEER

Iximeer uses a C++ core library for maximum compute performance, and interfaces directly with NVidia GPUs to provide the lowest possible on-screen latency. We use this compute power to allow the user to work with a combination of embeddings and NLP (natural language processing). Iximeer offers a very rich set of visualizations, our wordwall was specially developed by us to quickly identify & show text differences between documents. There is exactly the right representation for every data type. By putting all data at the fingertips of our users and immersing the user in the data, we let the data tell the story. If we break that virtual low-latency experience because the user must switch between views or components, he or she will quickly lose track. So, we had to ensure that all data is always in view and design our data interfaces around that core concept

Technology we use.

Technology is the core of our business, but we could never have gotten this far with our technical details without standing on the shoulders of some giants. In our case we use well-known and industry-tested components from the open-source domain. We have chosen these technologies because they are fit for purpose and have the right level of maturity. Every component must pass our safety tests before being included in our stack.

We make intensive use of Deep Learning for our internally developed image and text processing logic. We support this through various AI platforms. In our choice, we always ensure that it fits well within our customer’s Microsoft Azure landscape. We ensure that data never has to be edited outside the security perimeter.

INTERNAL API’s

We carefully ensure that our platform remains modular and manageable. Internal APIs ensure that components can be replaced if necessary, simplifying our product life-cycle management.

Our web front-end does not exist without several crucial open-source elements. Our interfaces must be as responsive as our back-end systems, we carefully select the building blocks that construct the core functions of our platform.

AI – SYMBIOSIS BETWEEN HUMANS AND AI FOR AUGMENTED INTELLIGENCE

Ixivault is a visual scanner of data that makes ‘cluster labeling’ possible. By working with data in entire groups at the same time, we greatly accelerate the labeling process. Data, even without meta-data, i.e., dark data or forgotten data, is automatically sorted by content. Possible labels are calculated for each cluster based on the content, after which a domain expert chooses from these labels. The data generates the labels, and people choose from them.

In this way, well-labeled data increases the quality of the AI ​​outcome, with less risk of bias or unethical data use. And with less resource effort and costs. But the direct costs are also lower by wasting less expensive compute power on what later turns out to be worthless and uninformative data.

Synerscope uses augmented intelligence, a design pattern for a human-centered partnership model in which humans and artificial intelligence (AI) work together to improve cognitive performance, including learning, decision-making and novel experiences. We anticipate that this ‘human-in-the-loop’ approach will always remain necessary due to all kinds of risks associated with increasing use of AI, such as regulations, explainability, financial impact, privacy, reputational damage, ethical standards, and values.