"The Kinsol team is very strong technically and great to work with. They helped us modernize and support one of our key product suites."

Chris Palmer, Director of IT Intrinsyc Technologies Corporation

projects

What's on the road?

We've trained deep learning networks to determine the difference between cars, buses, bicycles - and more - using vehicle detection analysis.

Our projects have included a proof of concept system in partnership with Quartech for the Government of British Columbia and developing innovations in machine vision for real time traffic web cams to determine traffic intensity.

In these projects, dozens of real-time traffic cameras are continuously processed with machine learning algorithms which have been implemented in a scalable fashion on cloud computing infrastructure. Kinsol’s mastery of the latest developments in deep belief networks has made this possible.

In a similar traffic-related field, we have worked with Kapsch and Streetline to examine a number of data inputs, including traffic cameras, in-road magnetic sensors and smartphone data to predict parking. As with all our projects, we deliver solutions on time and on budget.





Machine vision, deep belief networks,
faster regional based convolutional neural networks

“Kinsol has been a trusted resource over the years for delivering high quality inference solutions to problems in a timely manner.”

Mark Noworolski, the CTO of Streetline Inc.

All those
emerging
technologies

You name it and we've done more than just hear about it: the latest machine learning and big data techniques in natural language processing, recommenders and computer vision.

Making excellent predictions about customer behavior in relation to specific products is essential to prevail in today’s marketplace.

We’ve worked with many companies to apply the latest in recommender engine technology. We use machine learning techniques with collaborative filtering (in conjunction with cross validation) to create productized algorithms for recommending purposes.

When we did this for GlobalWide Media, we provided a proof of concept algorithm using a static corpus of historical data that could be used to determine the performance of matches that occurred in the past.

With Ziploop and Tictalking we have gone further by using natural language processing to determine sentiment and interest and matching that with recommender engines. Nainesh Agarwal, former CTO of Tictalking says, “Kinsol built an innovative recommender engine for us based on natural language processing. They’re a strong group of scientists and engineers and a pleasure to work with.”





Big data, Cloud / Edge / Fog Computing, Applied research

“We were very impressed with the competency and professionalism of the Kinsol team. They rapidly prototyped a sophisticated data mining algorithm for us on time and under budget.”

Nick Holland, CTO of Go2mobi

Smart Appliances

From super intelligent ovens or parking sensors that detect your car, we know "embedded" & how to unlock artificial intelligence between embedded and the cloud.

Kinsol was working with complex, embedded technologies before ‘Internet of Things’ was even a term. We developed our own proprietary mesh network, SpiderMesh, anticipating the growth of internet wireless collection for massive amounts of data. SpiderMesh, is our low powered, reliable, 900 MHz mesh networking multi-hop radio stack.

The IoT allows data to be collected today, that can be harvested in future. We’ve specialized in data collection and harvesting including physically challenging situations – such as collecting data on vital statistics in brewery tanks. As James Anderson of Phillips Brewing Co. says, “The Kinsol Research team has been a great help with a variety of industrial automation tasks from machine level coding to data acquisition.  Kinsol has also adapted themselves to my work environment and my schedule.”

Our embedded skills from the world of IoT and our experience in bringing the buzz of AI into function code in edge devices is helping Smart Appliance companies develop next generation cooking devices.





IoT, Firmware & system level design, algorithm design, wireless & mesh networks

Contact