5 Questions with Anupam Sengupta

Imagine a hard hat that can detect, alert, and prevent hazardous industrial work-related incidents through the use of advanced technology. GuardHat CTO, Anupam Sengupta, and his co-founders created just that. 

GuardHat’s “smart hard hat” has the potential to redefine worker safety. Sensors in the hardhat continuously transmit data to a safety control center, enabling the system to pinpoint each worker’s location and detect dangerous conditions.

The challenge for GuardHat came with processing and analyzing vast amounts of incoming data from the hard hats. Our platform has assisted the startup with their processing needs for almost five years, managing massive amounts of data every day, with the goal of keeping workers safe and secure. 

In our interview with Anupam, we discussed the beginnings of GuardHat, the fascinating technology behind the revolutionary product, and his vision for the company in the next 10 years. 

What prompted you to create a smart hard hat? 
Prior to starting GuardHat in 2014, the co-founders and I were working in heavy industrial manufacturing. We operated the fourth largest steel manufacturer in the U.S., supplying auto manufacturers across the country. 

One of the key challenges we faced – apart from trying to make the operations faster and more productive – was making sure that our employees were safe. Safety is paramount in the industrial world, especially in heavy industries where the work environments are not exactly the safest places. 

If you look at OSHA (Occupational Safety and Health Administration) statistics, you can see that improvements in workplace safety have stagnated. American industry is making widgets better and faster with more efficient, cost-effective processes. The processes around safety; however, have been trailing and haven’t significantly improved for decades.

That was the problem that started it all. At that point, we were the end customer looking for a solution. We had a certain product feature set in mind. Unfortunately, we couldn’t find anything to fulfill it.

We decided to form this company because we thought we had an idea that was worth testing in the market. Five years later, GuardHat is a success. It is a product that even I sometimes wonder why it hadn’t come earlier. 

Essentially, what we built is an IoT or a connected device platform. GuardHat blends in hardware variables, as well as infrastructure components, including real-time positioning, data streaming, audio-visual communications, and a mass volume data ingestion from sensory systems.

What was one of the biggest challenges you faced in the early stages of developing GuardHat?
The biggest challenge for all small companies and entrepreneurs in general is that you are resource constrained in terms of skill sets and time. Time is the biggest resource restraint that we always have. 

When starting GuardHat, it was important to prototype products quickly and fail fast so that we could move on to another potential solution. This was an area where HPCC Systems proved to be very useful for us. We didn’t have to spend months trying something out, because we had a system where we could quickly kick the tires, check its capabilities, and understand its limitations. The quick feedback and self-contained system that HPCC Systems provided became a very large asset for us. 

How has HPCC Systems contributed to GuardHat’s success?
When we started five years ago, we were a four-member startup, sitting in a garage trying to chalk out the next big idea. GuardHat is a very large system, and it’s not just software. It’s software, hardware, and a lot of communication systems that come with it.

On the data front, I was looking for a system that would deliver scale and performance. Just to give you a sense of the data volumes we manage, every single device generates close to 20 megabytes every eight hours. Multiply that by three shifts, and you have close to 60 megabytes of data that need to be processed every single day from each device. A site can have anywhere between 100 and 1,000 devices, and this is just the sensory data. It doesn’t include media and other data that we can also handle. Obviously, processing that volume of data efficiently and reliably was of key concern.

We were also looking for low latency performance. Because it is a safety system, I cannot depend on an operation that is inefficient or slow. I need it here, and I need it now — by which I mean sub-second resolution. 

I was looking for a pre-packaged solution – you open the box, turn a few dials and you’re ready to go. That’s where HPCC Systems entered the equation. Within the first week of downloading the platform, we decided that HPCC Systems was for us. The overall out-of-the-box experience — installing it, getting the console ready to go, and looking through your documentation — truly surpassed all of our expectations. 

In your opinion, what sets HPCC Systems apart from other open-sourced platforms?
For me, the biggest thing that HPCC Systems offers is that it is completely self-contained. You have all the pieces that you need to make things happen. HPCC Systems also scales up very nicely, which has been a key component for our growth. 

We are still finding new use cases for HPCC Systems. Just to give you an example, we are now finally getting into the area of machine learning and A.I., and we are looking at HPCC Systems to process some of those tools for us. 

Another distinguishing factor of HPCC Systems is the level of documentation and assets around the product. The tutorials and documents the company provides are incredibly helpful, and I always know that if I have a question, the support team is just an email or call away. 

Ten years from now, where do you see GuardHat?
Within the last four years, we have evolved tremendously. When we started this journey, we were heavily focused on the safety use case in the industrial workplace. However, what we unknowingly created is an IoT platform that provided a connected worker experience. In the industry, we have a term for this — Industry 4.0. Industry 4.0 deals with virtualizing and digitizing operations, which provides a real-time view of what you do and how you do it. We have evolved into something similar to that vision. 

In 10 years, GuardHat as a system, as a platform, and as a company, will have an even wider range of capabilities. Ideally, our system and product will know exactly what’s happening to workers in real-time in any part of the world, providing a transparent enterprise that integrates with the rest of the operational systems and essentially becomes the backbone of any system. 

We also might be running HPCC Systems on the hard hats themselves in 10 years. It’s not that far away. Ten years from now, Flavio, I’ll be calling you to tell you that we’ve just run HPCC Systems off of my hat. How about that?

Want to hear more from Anupam’s interview with Flavio Villanustre, VP of Technology, LexisNexis Risk Solutions? Listen to the webcast to hear more about one of our most interesting use cases.