Be ready to iterate, and start back at the beginning, armed with new experience. — We took the feedback and used it to revamp the development process for any particular application, such that we can deliver multiple levels of results along a prescribed timeline. We are able to accurately promise and deliver as a function of the amount of time we devote to data collection, giving realistic expectations along that process.
As part of our series about young people who are making an important social impact, I had the pleasure of interviewing Damas W. Limoge.
Damas W. Limoge is a Senior R&D Engineer at Nanotronics, focusing on nonlinear system control and integration with computer vision and deep reinforcement learning algorithms. While at Nanotronics, he has led the development team for a novel inferential process control platform, giving AI the power to maximize productivity in a factory. Additionally, he worked extensively on the nHale™, a low-cost breathing device that was concepted, designed, built, and acquired emergency use authorization from the FDA in under 90 days for the treatment of COVID19 in traditional healthcare facilities as well as private homes. Previously, he studied at the Massachusetts Institute of Technology, focusing on adaptive control methods for advanced battery management systems. This culminated in a control-oriented model of a lithium-ion cell, and the development of a novel adaptive observer for parameter estimation of multi-input, multi-output systems. He has given talks at the American Control Conference, Embedded Vision Summit, and Computer Vision Festival.
Thank you so much for doing this with us! Before we dig in, our readers would like to get to know you a bit. Can you tell us about how you grew up?
Igrew up in rural Vermont, in a close-knit community of teachers, community organizers, farmers, builders, and givers. Self-sufficiency undergirds a lot of the daily activity, but not in a sort of heroic way, rather as a means of survival. When something breaks, you learn how to fix it yourself, either through toil or the help of your neighbor. The community fortifies in times of hardship, with a mode of smaller self-sacrifice for the betterment of the public.
Is there a particular book or organization that made a significant impact on you? Can you share a story or explain why it resonated with you so much?
In some ways, my life is divided into epochs that are defined by the books or media I’m consuming at the time, so this is a difficult question. The first that comes to mind as it relates to my work is Zen and the Art of Motorcycle Maintenance, which chews over the nature of quality in all aspects of one’s life. I started reading Zen as I was making my own solo motorcycle trip across the country, from Boston to Palo Alto, stealing a moment to decompress in my bivouac amid the wilderness, thinking the content would simply pair well with my daily activities. Instead, I was presented with ruminations on striving for Quality and the traps along that journey, through the lens of machine maintenance. Quality is abstract, neither objective nor subjective, and as such, exceedingly difficult to define in concrete terms for static, programmatic achievement. On multiple instances along my trip, my motorcycle broke down in some form: a leaky gasket here, a sputtering motor there; these were instances where I could delay the progress of my trip by bringing the bike to a mechanic or work to understand the problem and fix it, bettering both myself and the machine. My motorcycle ended up being in better condition by the time I arrived in California than when I left the East Coast.
You are currently a leader in an organization that is helping to make a positive social impact. Can you tell us a little about what you and your organization are trying to create in our world today?
I am the Senior R&D Engineer at Nanotronics, a company that is wholly concerned with the nature of improving manufacturing processes. In many ways, this can be achieved with our foundational technology, which can detect microscopic visual defects with an efficiency a human could never achieve. Our work takes these images and correlates them back to sensor measurements of the process, giving an AI agent control of making small, safe adjustments to the process to reduce the likelihood of errors. Furthermore, we have used these techniques in the development of our healthcare devices, leading to higher surety of the production process. Many expensive products are priced according to the cost of low-yield subcomponents, extending to healthcare, consumer electronics, and the automotive industry, to name a few. By improving the likelihood of high-quality outputs, we reduce manufacturing waste and decrease the cost of these products, making them more accessible to a larger market.
Can you tell us the backstory about what originally inspired you to feel passionate about this cause and to do something about it?
Excepting widely publicized automation efforts, manufacturing has been starved of fundamental innovation for the better part of a century. Automation, while increasing productivity, only replaces the intrinsic mechanical inefficiencies of human labor, without considering the more nuanced deficiency of human reactivity and data interpretation. Nanotronics is keen on targeting every level of the manufacturing stack for improvement, and in doing so will deliver a production line with lower waste and fewer mistakes. We extended this to the manufacturing of our healthcare devices to create a high-quality product to target an underserved consumer demographic in our healthcare market, namely those that cannot afford the more expensive treatments.
Can you share the most interesting story that happened to you since you began with your company or organization?
The most revelatory experience we have had in our research to this point was in our formulation of the learning problem. At first, we considered the data streams to be highly structured: we would impart human intuition at each stage to ease the learning problem. We ended up getting stuck in a local optima in which our algorithms performed well on training data, but could not produce the results we desired when executing for validation sets. Only when we loosened the reins on a particular portion of the training set, namely the initial interpretation using unsupervised learning on image data, did we see promising results on the more complex elements of the problem. Machine learning is capable of incredible achievements, and to know the right time to be structured versus abstract in origination is informed by, in some ways, the human learning process.
Can you tell us a story about a particular individual who was impacted or helped by your cause?
We have begun working on pilot projects in many fields, but one in particular, a manufacturer of household medical disinfectants, is changing the industry through distributed production with smaller-scale equipment. The nature of low-volume production per reactor means the lessons learned from a long history of manufacturing can no longer be applied. After a few months of data collection, we are able to help predict identifiers of anomalous production behavior and infer the state of processes that were previously unmeasurable. These can be used to further improve production yield and fortify the viability of the business model. As we collect more data, we will be able to not only correlate process measurements to quality output, but wield those correlations to further improve yield and their bottom line.
How do you define “Making A Difference”? Can you explain what you mean or give an example?
Making a difference means using one’s skills or capability to enable another to do so. In this way, making a difference is a feedback loop that betters community through multiple levels of action. The platform we have built has an emergent property that gives an opportunity to small-scale, disruptive technologies to leapfrog years of development to grow into viable, accessible methods for mass production. This lessens the capital investment requirement for our peers and brings their products to those that need them at even faster rates. We have demonstrated this with our own manufacturing processes, delivering critical life-saving equipment for a fraction of the cost of other manufacturers.
Many young people would not know what steps to take to start to create the change they want to see. But you did. What are some of the steps you took to get your project started? Can you share the top 5 things you need to know to become a changemaker? Please tell us a story or example for each.
Driving change is a seemingly alchemical process that is at the nexus of creativity and perseverance. Distilling that change to five items might look like:
- Identifying a real pain point to provide a viable solution. — We knew we wanted to connect the feedback loop from our inspection back to the control of factory processes, but it was only after we had settled on improving the reliability of additive manufacturing as our proof of concept did the fundamentals of the technology start to take form.
- Let the problem drive the approach, and not a desired technology. — We began forming a black box stack that could solve any problem, including ours, but required significantly more data than could be collected in a reasonable time frame. We decomposed that black box into elements that could remain ubiquitous without a hunger for immense data collection, and conditioned other parts of our problem to leverage human intuition to make progress while limiting data intake.
- Focus on the real world application of one’s change, instead of the laboratory settings. — After getting great results in our lab, being able to make inferential decisions to improve the consistency of the quality, we were not satisfied to simply rest on that achievement. Instead, the next step was finding viable partners to use this technology in the real world to show the extension of our conceptual work.
- Get feedback from peers and real users of your solution. — A peer review of our work, as well as the feedback from our strategic partners, was invaluable. We were able to gauge the perception of our approach as it related to the state of the art, while grounding ourselves to the real experience of users and their needs.
- Be ready to iterate, and start back at the beginning, armed with new experience. — We took the feedback and used it to revamp the development process for any particular application, such that we can deliver multiple levels of results along a prescribed timeline. We are able to accurately promise and deliver as a function of the amount of time we devote to data collection, giving realistic expectations along that process.
What are the values that drive your work?
Above all, our team strives for quality: quality in our products, but also quality in the products we enable. It is this search for quality that defines all human endeavor, and our only difference is we do so in explicit terms. We couple that with a focus on maximizing access to our enabling technologies, not just for those with means, but those that need them.
Many people struggle to find what their purpose is and how to stay true to what they believe in. What are some tools or daily practices that have helped you to stay grounded and centered in who you are, your purpose, and focused on achieving your vision?
Allowing oneself to be entirely subsumed by just a segment of the human experience is limiting, possibly leading to a dangerously parochial view of the problem landscape. It may be counterintuitive, but I find a healthy balance of discipline and leisure to be the most grounding practices in my life. Allowing myself to be humbled through learning outside of my expertise and sampling the vastness of talent in our world garners a refreshed approach to focusing on the problems that sit before me every day. It also reminds me of those that come from different backgrounds that may be underserved by the dominant voices of an industry.
In my work, I aim to challenge us all right now to take back our human story and co-create a vision for a world that works for all. I believe youth should have agency over their own future. Can you please share your vision for a world you want to see? I’d love to have you describe what it looks like and feels like. As you know, the more we can imagine it, the better we can manifest it!
I see a world that is inevitably trending toward extreme specialization and the elimination of human physicality in mass production. What will remain is bespoke, artisanal production where the human element is implicitly what gives that work value. This will not eliminate our want or need for widely accessible products, and as such, we will enter a stage in which a large portion of the labor force will be disenfranchised. This has already begun, and it has bred accusation, defensiveness, and the feeling of betrayal, while also endangering the lives of those the system does not serve. Ultimately, we will need to address the nature of value and health being tied exclusively to revenue-driving productivity, in that most humans will not have access to this type of work opportunity. Instead, we will have a world in which our factories are mostly driven by machine interpretation and execution, allowing humans to pursue the type of work that is bolstered by their innate interests outside of a profit-driven model of labor and, subsequently, survival. A manufacturing model that is so efficient it drops the financial barrier to expensive medicine or microtechnology for markets that would otherwise be ignored for a lack of spending power. I hope to enable a world that returns human labor back to the work that inspires the individual, while necessitating the societal and fiscal policy that must accompany such a transition for a healthy, loving population.
We are powerful co-creators and our minds and intentions create our reality. If you had limitless resources at your disposal, what specific steps would take to bring your vision to fruition?
With unlimited resources, I would enable all factory production to be driven, and enabled, by an interest in optimized quality, thereby lessening the focus on profit margin as a function of output yield. This cannot be achieved by technological development alone and must include a political effort to give buying power to the masses even without income tied directly to labor. A world in which a personalized output, however low or high that may be, is sufficient to grant access to all of the benefits of modern technology, including effective healthcare. We have the tools to give so many a happy and comfortable life, truly in unprecedented terms, but in many ways the anachronistic structures that previously enabled development underserve the children of those who built them. This must change because, simply put, our survival depends on it.
I see a world driven by the power of love, not fear. Where human beings treat each other with humanity. Where compassion, kindness and generosity of spirit are characteristics we teach in schools and strive to embody in all we do. What changes would you like to see in the educational system? Can you explain or give an example?
First and foremost, the accessibility of education is the most significant change we need. I mean this in a holistic sense: learning and development is an integrating function from the moment we are born and continues throughout our lives. Mothers need support during pre-K childcare, children who went to public school (like myself) need to be given the tools to compete with their peers at private institutions, postsecondary education without the specter of debt should not just be a privilege for those with financial means, and the choice to return to school at any point in one’s career, whether following a conventional educational path or not, should be viable and encouraged. This is a topic that is particularly close to my heart, as I have family members who have dedicated their lives to widespread access and preparedness for continuing education in the public domain. We so frequently see the culmination of the story of those left behind in the educational system, simply because they were not surrounded by the opportunity to better themselves through truly equal means.
If you could tell other young people one thing about why they should consider making a positive impact on our environment or society, like you, what would you tell them?
My life goals are about increasing the radius of positive influence. Consider a vector drawn from the center of your being, encircling a sphere of those you support and whose support you are privileged to have. It starts with yourself: we must love and appreciate ourselves, while maintaining a critical eye for the ways we are not having a positive effect. Then comes our close family and friends, those that are tied to us by prolonged experience. We must support them in the way we hope to support ourselves and can only achieve that through a fortification of our own being. Next, we have our schools, communities and workplaces. Then our cities and our societal infrastructure. And, perhaps the most challenging, is to eventually support our global community and the human species. Each level both enables the next, while being enriched by the continuity of that support. It is a process of learning and understanding that which enables us to do more. When we strive to make a positive impact for those around us, we unexpectedly sure up our own value system and sensibilities.
Is there a person in the world with whom you would like to have a private breakfast or lunch, and why? He or she might just see this, especially if we tag them. 🙂
I admire the thought work being conducted on the origin of training data in machine learning, namely as it relates to race and accessibility. There is so much prejudice in the type, shape and volume of data selected to train models, and the scientists and engineers responsible for that training imprint conditioning shaped by societal norms on that choice. One organization that is doing great work in developing thought around this problem is Data for Black Lives, led by Yeshimabeit Milner. I am sure I would learn a lot from just one conversation with her, and I would hope to take that back to the work I do.
How can our readers further follow your work online?
To see what we are up to at Nanotronics, you can follow us on Instagram, LinkedIn, and Twitter. Our CEO, Matthew Putman, has a thought-provoking and entertaining podcast called Utility + Function where he interviews subjects across the arts, technology, physics, philosophy and more on their individual journey toward fulfillment and success. It can be found on Spotify, Apple, and Google Podcasts.
This was very meaningful, thank you so much. We wish you only continued success on your great work!