The Future of Edge Computing: Distributed Computing.
Litmus Automation’s Lead Software Engineer, Dmitri Toubelis on his career and thoughts on the future of edge computing.
Dmitri Toubelis is Litmus Automation’s Lead Software Engineer for LoopEdge. We took a moment to ask Dmitri a few questions on his previous experience within automation, his current work on LoopEdge and his thoughts on the future of Edge Computing.
What is your professional background?
My career started in the Soviet military, flying strategic bomber planes during the Cold War. My specialty was in communications and electronics. I later switched from working with electronics to computers. My first paid job was in 1986 or 1987 working specifically within industrial automation.
Wait, you flew strategic bomber planes ?
Yes. A great story for my kids.
Interesting! What got you interested in electronics and computing?
As a kid, I worked endlessly on electronics as a hobby. There were military schools that specialized in electronics and automation. My uncle was an airforce colonel, and my parents suggested I join the airforce so I wouldn’t be digging trenches.
While in the airforce, my study and work habits were similar to when I was younger; I did whatever I was required to do, but I did it quickly so I could free some time for more tinkering on electronics. I worked eight hours a day at my job, and I worked another eight tinkering on the side. Lots of learning came out of this tinkering. I never had to study for exams!
Eventually, my interest in electronics transitioned to computers. For example, I used to sneak in to the computer lab to write programs to simulate assignments on circuitry and electronics. We were supposed to do this by writing out all the formulas on paper, but I thought, why not automate it and make a buck doing these assignments for the rest of the group? So, I would do thirty assignments under an hour using a program I created.
I wasn’t part of the faculty that studied computers, so one of the teachers asked what I was doing in the computer room for so long. I showed him what I made and he couldn’t believe his eyes. He didn’t know how to do what I did.
How did you hear about Litmus Automation?
I decided to get involved with start-ups about seven to ten years go. During those years I worked with a few companies to establish the foundation of their technology and business, until Litmus Automation CEO, Vatsal Shah, found my resume and called me directly. After talking with him, I knew joining Litmus Automation would be a great opportunity to use my skills in technology and automation.
What did you and Vatsal talk about?
By that time I had already gained a lot of experience in how to judge which companies would be successful and which ones had very little chance despite any help I could provide. Vatsal was one of the few people who actually understood what makes a business successful, and for me, it’s rare to find people like Vatsal who are smarter than myself. Soon after my call with him I joined the Litmus Automation team. I knew that Vatsal was smart about industrial automation, and if we joined our efforts we could do great things together.
Can you describe your role at Litmus Automation?
When I joined Litmus Automation I was Head of Engineeringfor LoopEdge. From the beginning, the goal was to use best of breed technologies to make LoopEdge competitive and successful. We had to build everything from scratch using Jira, Confluence and several other development tools. We set up proper methodologies and workflows, distributed responsibilities, developed an architecture, and brought together an amazing team. It’s not about just building a product, but building a great team. I see my role as providing guidance on how to move LoopEdge development in the right direction to be successful for the next five, ten, even fifty years. We make decisions collective and collaboratively, but I steer the team using my lens of previous experience with technology and manufacturing startups, as well as a vision for the product. Litmus Automation is growing rapidly, and we have to make sure the growth is controlled so we don’t run into deficiencies in our organization.
What are your thoughts on the future of industrial automation, specific to edge computing?
IoT, and specifically Industrial IoT, was hyped for so long it almost felt like a work of fiction. For years the industry said – what if we could connect and control devices – there would be no boundaries to what we could do. However, no one created anything tangible. Litmus Automation is one of the only companies actually doing something tangible with several real-world applications. We have done a lot to get our foot in the door in industrial automation and edge computing, but the future lies in actual distributed computing at the edge, artificial intelligence and machine learning. We are enabling those technologies on our platform to get into the really exciting stuff.
Distributed computing excites me most, although I don’t think anyone understands what it really means. Most think, “Okay, we have this gateway edge device with spare computer power, lets run some applications on it”. However, because the device is relatively small and not powerful enough alone, the true potential of edge computing lies in combining several devices and performing distributed computations that could not be accomplished on a single device. As the number of devices grows within a manufacturing company, distributed computing will become a more obvious approach. I want to take LoopEdge in this direction; investing in distributed computing rather than just edge computing.
I believe this type of distributed computing is the future. After we move towards distributed computing and distributed workloads between gateway devices, the system will work seamlessly without the end user knowing how the workload is done. That is, in my mind, the power of edge computing that nobody has tapped into yet. If we continue on our path, and our leading position, the potential is unlimited.