Matthew Kavanagh is an Algorithmic Engineer at the Hyperganic Group, focusing on the deployment of AI tools to improve the accessibility and speed at which complex engineering designs can be made using the Hyperganic software. Matthew specialized in biomaterials, nanomaterials, and 3D printing when obtaining his master’s degree in Materials Science and Engineering from Imperial College London. His thesis focused on the mechanical applications of architectured metamaterials fabricated via additive manufacturing. in 2022, Matthew joined Hyperganic in Singapore and carried out research on the application of machine learning in the design of additive manufacturing space. His latest work is centered on the upcoming release of a ‘Text-to-3D’ AI tool that aims to enable users to describe design alterations using natural language when integrated with Hyperganic’s software. Matthew will be speaking at the upcoming virtual event focusing on AI/ML in healthcare 3D printing.
When was the first encounter you had with 3D printing? What was that experience like? What were you thinking at that moment?
Matthew: My first encounter with 3D printing was during a 3rd year university group project. In teams we had to create a marble run as a design challenge that incorporated coding motors and sensors to detect and move marbles along our section of the run. The structure used to build the run itself was designed in CAD and then fabricated using 3D printers. This project was really eye-opening for me as it was the first real experience in my degree that promoted creativity much more freely than the predominate focus on scientific theory.
Having studied a mixture of physics, chemistry and biology across my Materials Science degree I was immediately drawn to the possibility of utilizing this additive process to enable the creation of new ‘smart materials’ or structures that had properties not seen before in nature (meta-materials). This was a huge spark of inspiration for me as the reason I chose my degree was driven by my interest in smart materials and the discovery of new ways of organizing matter to improve performance across scientific and engineering disciplines.
I had always known about 3D printing prior to any project I did but, getting hands on with it inspired me to pursue more knowledge in the field. In the final year of my Meng degree, I chose to take a course outside of my host department on ‘Design for Additive Manufacturing’ as well as selected a thesis project that involved 3D printing in order to develop my understanding of the technology alongside the latest scientific literature.
What inspired you to start your journey?
Matthew: What really inspired me to get involved in the 3D printing industry was the content I explored during my master’s thesis that focused on mechanical architectured metamaterials that were printed via additive manufacturing. Learning about the unique properties of the structures that could only be made via complex design and a bottom-up manufacturing approach got me excited by the prospect of potential applications.
From my research, I felt that as the field’s understanding of the complex relationship between microstructural evolution, printer technology and printer parameters developed, we would soon be able to produce parts that massively out-compete the performance of traditional engineering objects.
Having specialized in biomaterials for 2 years I was also heavily exposed to bio-printing for the creation of scaffolds and patient specific implants. There is something extra compelling when you study how these types of scientific breakthroughs can have profound improvements on people’s quality of life, especially when you are able to imagine yourself with similar aliments or injuries in the future and the level of technology you would desire to be put into your own body.
My holistic exposure to 3D printing in my final years showed me just how diverse it was and the plethora of industries and research that it could enable progress in. This fueled my motivation to join a company working to develop related technologies to speed up or inspire a transition towards AM globally. Once I found Hyperganic and saw what was being attempted, I was immediately drawn to the possibility of finding a more accessible way for people to engage with the complex designs through coding algorithms.
Who inspired you the most along this journey in 3D printing?
Matthew: I would say my inspiration for pursuing 3D printing comes more from my deep fascination of science in general as opposed to any particular person. I see it as a technology that may at some point enable a sustainable cyclic manufacturing industry which also enables engineering to proceed to the next level of complexity.
It is common in science fiction to see machines able to create any object desired and there are no theoretical grounds that prevent the possibility of printing perfect objects from the nanoscale to the macroscale. Whilst there are huge challenges that keep such a machine a while from existing, it seems possible that ‘refabricators’ of this kind could be created in the future and this is something that greatly excites me.
The first book that really put me onto the path of pursuing cutting-edge scientific technology was Michio Kaku’s “Physics of the Impossible”, where he dives into what the physical existence of science fiction concepts would or could look like. Having read this I started to see just how unbounded science is and that we are still very far away from understanding all its intricacies.
What motivates you the most for your work?
Matthew: My biggest motivation has always been to improve the quality of life of people around the world. To me, this (as well as the pursuit of knowledge) has always been an ultimate goal of science, yet one that doesn’t always seem to be the predominant driving force behind technological development.
I also think that we need more people to be pushing forward science and engineering, which I believe will come from lowering the barrier to entry into the field with more intuitive, accessible tools that enable complex objects to be designed and tested with less initial knowledge required. Ideally, everyone should be able to experiment and produce objects related to their special interests without needing such in-depth knowledge of theories across multiple scientific disciplines.
Another big motivation for me comes from my own passion for knowledge, I have always been in awe of the desire for understanding shown by philosophers from antiquity to the modern day and I see scientific developments as a large (but not complete) piece of that jigsaw. Whilst there is still a lot to be understood in the large world of physics as quantum mechanics develop, I think additive manufacturing gives us a key window into a clearer picture of the emergence of properties as matter and structures progress from nanometers up the hierarchical scale.
What is/are the biggest obstacle(s) in your line of work? If you have conquered them, what were your solutions?
Matthew: I think one of the biggest obstacles when trying to develop software that enables users complex design freedom is finding a good balance of providing tools with encoded engineering knowledge without encapsulating so much that the users feel out of their depth trying to understand what tool is doing for them. Whilst this will always be an ongoing solution, with constant improvements, it is important to enable people to come in with their own expertise and simply remove repetitive engineering tasks so they can focus on the development of the design itself. It is also imperative that we close the design loop so that with a few clicks, fully optimized parts can be produced.
Another big issue that I think many of us are facing at the moment is the safe and carefully considered deployment of AI/machine learning tools to speed up the development of science and engineering. Again, this is not something to conquer but it is vital that there is constant progression and rapid evaluation throughout development and release cycles.
Finally, a large obstacle we have faced is how to approach the re-education required to create a shift in mentality from traditional CAD to computational design. Algorithms are much faster and can automate many processes, but finding a balance between enabling maximum design complexity whilst teaching easy accessibility is a tricky problem that needs to be overcome to entice people to get excited by new possibilities in this field.
What do you think is (are) the biggest challenge(s) in 3D Printing/bio-printing? What do you think the potential solution(s) is (are)?
Matthew: For me, the biggest limitation in the additive manufacturing field is our understanding of how exactly these complex designs form across all the different types of AM technologies as well as the infinitely unique printer/process parameters that are used in the fabrication. Whilst polymers are slightly easier to study, the formation of metal microstructures remains a huge task given the effects that even the most minute compositional differences can make on the properties of the finished print. There are similar difficulties in the bio-printing field where not only is the design and quality of the print important but also the impact all of it has on the material’s interaction with their target cells.
I think it is clear that machine learning will play a large role in speeding up this process of understanding by finding smaller latent spaces from the infinite possibilities to try to point us toward the factors that play the biggest roles throughout these processes and the connections between them.
Whilst it is easy to say that 3D printing is still an emergent field so more money needs to go into its development and more years of studies are required, I would still prefer to see the open sourcing of knowledge first. I think tackling these huge engineering challenges will involve deep learning models that are able to train across global AM data to find prominent trends faster than before and in a way that enables more people to use the information.
Obviously, there are problems that come with such an open-source policy but given the pressing times of climate change and general global destabilization, I think collaboration across research disciplines, industry, and countries is really what is required for us to keep pace with developments, especially as AI continues to progress at an alarming rate.
I think CERN has been a great testament to continent-wide research that is carried out by those with a passion to drive forward knowledge as opposed to anything else. It would be nice to see more large-scale projects like this in the additive manufacturing field to really start pushing the bar on what is possible for engineering in the near future, and I see no reason why a project as large as CERN couldn’t be created for engineering developments as opposed to theoretical physics.
If you are granted three wishes by a higher being, what would they be?
Matthew: I think the first wish I would ask for would be perfect memory and recall, information is so easily lost by the brain, and we are all fallible to the imperfection of memories and experiences. This would enable much more stable development of concepts across a lifetime and probably would enable faster connection of complex ideas.
My second wish would be to perceive or experience all forms of energy. Whilst we as humans sort of already do this, I think the value of being able to “see” all levels of the electromagnetic spectrum for example, or “seeing” waves of vibration would open new ways to understanding the processes of the natural world.
My final wish would probably be something more standard like a superpower to tie these other wishes together. For this reason, I’d wish to have similar abilities to Marvel’s Ant-Man. Being able to traverse all length scales from tiny to large and even the quantum realm would grant you unprecedented access to all the processes in the world and when combined with the other wishes, would enable unparalleled knowledge which could be converted into technologies that improve global quality of life.
What advice would you give to a smart driven college student in the “real world”? What bad advice you heard should they ignore?
Matthew: The best advice I can give is to never be afraid of asking “stupid” questions or asking for clarification on something even if you have already covered it and think you should know it. Understanding comes from repetition and being afraid to clarify even the most basic concepts can build shaky foundations in the future.
Following on from the last point, always be willing to re-read or re-learn something. It is very unlikely you fully grasp the entirety of a concept on your first try and as you learn more, the way you process information also changes.
Being flexible and tolerant is also something I have found to be key. There have already been many moments where beliefs I felt strongly about have been challenged or I have stumbled upon contradictory information. The ability to accept and take on different views and use them to constantly develop and finetune my working model of the world has paid dividends. Being too engrained with one angle of the world distorts your experience and only enables you to process information that conforms to it.
In terms of bad advice, I have been given that I would ignore – never believe you must be in a certain situation/location to achieve something in life. That could mean you feel the need to be at an academic institution to develop scientific breakthroughs or in a huge global corporation to have a big impact on the world. Everyone’s journey is different and what works for you might be what works for everyone else, but it could also be something that has never worked for anyone before.
My final piece of advice is to never be afraid to be out of your depth or to fail. All the best things in life are on the other side of fear and failure. Take chances, learn from your mistakes, and never stop challenging yourself in whatever you do!
Related Links:
Advanced Visualization in Healthcare: AR/VR/MR (Course on Demand)
Artificial Intelligence in Healthcare 3D Printing (Course On Demand)
Interview with Kerim Genc: The Power of Artificial Intelligence and 3D Printing
Comments