The Silicon Review
SWARM Engineering is a creation of our times; the company combines the latest Machine Learning and Operational Research algorithms with an advanced multi-agent approach to solve complex problems in highly disruptive environments, such as a supply chain in a pandemic. Ten years ago, humanity couldn’t have solved many of the challenges we tackle today because AI research wasn’t as advanced, and most organizations lacked the necessary data to train machine learning systems. Now, not only is the technology ready but so are organizations – especially those in agri-food, who are going through a digital transformation and need to find ways to become more efficient and effective while increasing sustainability. The time to do this with pen and paper has gone, and simply relying on a flashy computer system won’t cut it either. You need a human-machine partnership. Humans augmented by cognitive AI is the way forward. And SWARM Engineering is here to deliver exactly that.
Founded in 2016, SWARM is a Software-as-a-Service platform that uses a next-generation cognitive computing system designed to tackle issues in the food supply chain and logistics such as load planning, inbound logistics, product blending, and pricing. The SWARM platform is structured around a multi-agent approach that can utilize a curated market of optimization and machine learning algorithms. SWARM provides an easy way for business users to define problems and rapidly match this to a solution without any software coding or knowledge of advanced AI or machine learning.
We recently interviewed SWARM’s founder and CEO, Anthony Howcroft, to learn more about the company and its innovation. In the interview, the executive also talked about the company’s future launches. Read on for the excerpts from the interview.
Q. What inspired you to establish SWARM Engineering?
Before SWARM, I worked at Microsoft, leading EMEA for big data platforms. We helped lots of organizations that were looking for actionable insights on their data, isolating issues and highlighting where remediation was required. I kept wondering if we could do more. Shortly after leaving Microsoft, I was in a casual conversation with the CTO of a division of GE, where we began discussing what it would look like if there was a Facebook for machines. I realized that if machines could collaborate with each other directly, we might solve problems before they became issues. I began exploring multi-agent systems and contacted some academic experts on machine learning, and that was the spark that ignited SWARM.
Q. How have you evolved with time? What makes SWARM unique?
Early on, we were building tools for data scientists. Very quickly, though, we discovered that most of our target customers had very few people with that expertise, and they were overwhelmed with projects already. We also noticed that many of the problems were much easier to define than to solve. So we set out to separate the problem from the solution and elected to make the problem or challenge (as we call it) definition easy enough for business users to perform without any knowledge of AI, ML or math. That separation has a huge, positive impact – such as the ability to switch in new solutions rapidly as algorithms improve. The way we have achieved it is, we believe, pretty unique.
Q. How is SWARM Engineering solving the challenges that plague food supply chains?
One bite at a time! We start by identifying a process that takes people too much time and has a high value impact. Then we define the ‘challenge,’ and automatically match it to a solution in our curated library, or source a new algorithm if necessary, and then rapidly deploy. Then we move up or downstream to fix the next big challenge in the customer’s supply chain. The days of big-bang implementations are long gone; you must get to business value almost immediately.
Q. Could you tell us about your offerings in brief?
We offer a SaaS solution that solves critical food supply chain processes such as maximizing yield, improving inbound or outbound logistics, balancing supply and demand, load planning, or network optimization – so you can decide which facilities to open or consolidate. The savings are typically in the millions of dollars range.
Q. How important are Algorithms and Machine Learning to your SaaS platform?
Algorithms and Machine Learning are at the heart of our solutions, but the business users don’t need to interact with them directly, because we offer no-code AI. We believe human operators should focus on the goals, scope and constraints of the challenge, and do that in clear business language. They don’t need to learn about Bayesian probability, Markov decision processes or deep reinforcement learning. Our algorithms are all open for inspection though, as we know it is critical for companies to avoid bias of all types, and nobody wants an impenetrable algorithm. Ultimately, algorithms and ML have huge value and we unlock those benefits. Our aim is to democratize AI.
Q. Tell us about the company’s work culture. How do you help your employees grow?
We are a startup, remote-first, and that means people need to wear many hats, and are encouraged to step out of their comfort zones. It’s a tremendous way to learn and grow. We share information across functional boundaries and hold a daily video standup meeting for everyone, so people get a more holistic view of the business. If mistakes are made, we learn from them, but we look at the process and don’t assign blame to individuals. We also share the good news and recognize great performance.
Q. How important are partnerships to SWARM Engineering? Do they play a pivotal role in what you deliver?
Partnerships have not been pivotal to our journey so far, but that has changed in 2021. This summer, we hired Cliff Currie, as VP Business Development to prepare for the launch of our next generation SWARM solution engine. Cliff and I have worked together in the past, and it was great to be able to bring someone onboard who I knew could have such a fast and strong impact. We now have a handful of very interesting OEM partnerships in development and made the first announcements this summer around an exciting partnership with Microsoft. We are also working on specific AI Research & Development projects with two academic institutions, which will be announced this fall. There is more to come, too – so watch this space!
Q. How did the Covid-19 pandemic affect SWARM and its operations?
The pandemic didn’t alter our platform or business model, in many ways it was exactly the type of challenge SWARM was designed to meet. What Covid-19 highlighted were elements of the food supply chain that need to adapt, and it also validated our approach for optimizing in disruptive environments. We believe the pandemic simply accelerated several trends that were already visible in agri-food; more home delivery, a focus on healthy eating, better risk management across suppliers, and so on.
Q. Do you have any new services/products ready to be launched?
Later this year, we have an exciting launch planned for a new SWARM application, which will make it far easier for any business user to define a challenge and match it to an advanced solution. It has the potential to transform the way organizations look at solving problems, and we think it can have a huge impact on sustainability, reducing both emissions and food waste, while simultaneously making dramatic cost savings.
Q. What does the future hold for your company and its customers?
We envision a future where food is safe, healthy, and affordable for consumers, while providing a balanced income to everyone in the food supply chain. The supply chain of the future will be able to take disruption in its stride to continually adapt and learn. Humans will manage and control the goals of the supply chain, while AI makes and executes the operational decisions to achieve those goals. We believe SWARM will be embedded in future value chains, automatically solving problems and giving end users abilities that would look like superpowers to somebody today – precognition, telekinesis, adaptability, mastery, telepathy – what could you achieve if you had those powers available?
Anthony Howcroft, Founder and CEO
Anthony grew up in a rural farming town in England and worked in operations at Kraft Foods. Subsequently, his career was spent working for a mix of corporates and startups, such as Texas Instruments and Microsoft. Anthony studied Creative Writing at the University of Oxford and is a prize-winning author of fiction. His first non-fiction book, Questions – A User’s Guide was published last year, where it entered the Amazon best seller’s list. He lives in California and frequently leaps into the air after mistaking dry sticks for rattlesnakes.