Privacy Preferences

When you visit any website, it may store or retrieve information through your browser, usually in the form of cookies. Since we respect your right to privacy, you can choose not to permit data collection from certain types of services. However, not allowing these services may impact your experience.

  • For the demand-oriented design of our website we create pseudonymous user profiles with the help of Google Analytics. In this way we are able to recognise returning visitors and count them as such.
  • These technologies enable us to integrate external services such as video sharing into the website.

Intelligent edge processing for better AI-models

A smarter, efficient selection of information from sensors - at the edge - delivering higher AI-model accuracy at lower latency on low power hardware.

What we do

Edge AI Lifecycle

Edge Cycle

WHAT WE DELIVER



   Teraki’s helps to build, train and improve sensor-data driven AI-models (IoT-algorithms).

   Teraki increases the accuracy rate of AI-models with an additional 10-30%. This additional reliability results into increased business value and better products.

   With Teraki the application runs 10x faster and consumes 10x less power.

   For high volumes of vehicles or IoT devices, the Teraki Platform automates the training for AI-models and improves the accuracy during their lifecycle.

HOW IT WORKS



   Edge: Teraki embedded client selects and encodes the essential information in any signal.

   Via the DevCenter customers can configure and visualize the performance of the trained models on the Platform.

   Cloud: The processed data is exported to the customer cloud to be used in customer’s AI-stack.

   The Teraki Platform provides APIs to ingest, train and deploy AI-models.

Teraki inside:

TERAKI PLATFORM
Ingest data, train and deploy models at scale

   Train, manage & deploy models

   Collect vehicle data

   API provisioning for software developers

EDGE SDK
Embedded software

   Low power hardware

   MISRA and AutoSar compilant

   Easy implementation via wrappers

Benefits.

Core benefits our solution provides.

Accuracy

Teraki’s AI powered pre-processing improves the accuracy of sub-sequent AI-models by 10-30%. Achieved by efficient edge processing for the extraction of high-quality data (without noise) in normal and difficult conditions.

Latency

Teraki’s algorithm reduces latency by 10x for detection algorithms, typically 10-30 ms latency on ASIC. This enables real-time operations.

Power consumption

Teraki’s pre-processing consumes minimal power and reduces CPU footprint. The energy consumption for AI model is reduced with factor 10x. Efficient power consumption is essential for longer mileage range of the car/drone/robot.

Lightweight embedded

Teraki’s pre-processing software is designed to operate in embedded devices and SoCs. It can run on various operating systems and is provided as a code-wrapper. The software achieves 4x -10x less data needed to train algorithms; hereby creating a suitable environment for complex AI-models to be operated on embedded hardware.

Continuous improvement

Teraki supports a continuous learning loop for customers to train and update pre-processing and their AI-models at the edge. The Platform enables customers to (re)train their own ROI/TOI models; refine the next data collection; and subsequently deploy more accurate AI-models. This process is automated to rapidly train new AI-models as well as to support AI-models that run in high-scale production.

Investors and recognition.

We received awards and are supported by:

Blog.

Latest Blog Posts

Importance of Last Mile delivery

The last mile is often the most expensive (nearly 1/4th of total delivery cost) and most time-consuming segment of the logistics. Delivery robots are increasingly viewed as a potential option for last-mile deliveries in the not-too-distant future.

Best-in-class AI-chip meets best-in-class edge-AI

safety of automation gives the need to operate deterministically with low resources such as CPU, RAM and power consumption at the edge. This gives the opportunity and challenge of making an efficient hardware and deploy algorithms that can run on such low resources and produce accurate results in real-time

Increasing AI-model accuracy and training efficiency through region- and time-of-interest techniques for edge applications

Despite the maturity of many models, they still require substantial computational resources. New AI applications need to go through training and testing phases. For these phases, data must be recorded in relevant environments to capture situations of interest required to refine the application.

Contact.

Contact us at info@teraki.com

For directions to our office, please follow the link below Offices