I recently attended DeviceTalks in Boston. Two days, 1,600 of the industry's sharpest minds, over 100 speakers, and hundreds of conversations. A great event with people who are helping shape the MedTech conversation.
What struck me was a consistent theme running through all the sessions: technology and MedTech are evolving at an eye-opening rate. Engineering, design, and manufacturing will only get more complex, more integrated, and more demanding.
At HiArc, we believe healthcare needs to keep innovating and integrating new technologies to better serve patients. An excellent example of that was the closing keynote at DeviceTalks.
It was Noland Arbaugh, the world's first Neuralink user, demonstrating live neural control from the stage. No mouse. No keyboard. Just thought, translated into action, in real time. If you needed a single image to capture where medical technology is heading, that was it.
Here's what stood out to me, and what I think it means for anyone building complex medical devices right now.
AI came up in almost every conversation, but not in the panic-about-jobs way the headlines suggest. The mood was pragmatic. Teams are finding genuine value in faster documentation, sharper data analysis, and tighter design iteration. That's real progress.
But the consensus was equally clear about where the value ends. AI cannot run your product development program. It cannot build your prototype or put your device into manufacture. No algorithm signs off on a first article inspection. No model manages the relationship between a design change and a supplier's process capability.
Implication: The teams that move fastest in an AI augmented world won't be the ones who automate the most. They'll be the ones who are clearest about which decisions still require deep human expertise, and who they trust to make them.
New England is still the heart of U.S. MedTech. In particular, Boston's MedTech ecosystem continues to grow steadily. According to MAssBioEd, Massachusetts alone employs over 143,000 people in life sciences, ranking number one for R&D talent in the US for four consecutive years. The density of expertise, institutions, and decision-makers here is unmatched.
HiArc is headquartered at the center of that corridor in New Hampshire. At DeviceTalks Boston 2026, three of our engineers, Tony Lacroix, Tushar Patel, and Sam Sarette, took to the stage alongside experts from Medtronic, and MIT. We're not commentators on this ecosystem. We're embedded in it.
Implication: When something goes wrong in a complex program, and it will, the partners who respond fast are the ones who are geographically and relationally close. Proximity isn't a nice-to-have. It's part of what makes engineering programs work.
DFM and DFX were recurring themes throughout DeviceTalks, and the message is becoming more urgent every year: too many technically brilliant programs still stall, slip, or exceed their budgets because manufacturing considerations weren't addressed early enough.
Material selection, tolerancing, process capability, automation readiness, yield predictability, these are not decisions you make at the back end of development. By the time you're fixing them, you've already spent the money and lost the time.
Implication: The programs that reach commercial production on schedule treat manufacturability as a design constraint from day one, not a phase-gate at the end. If your development partner isn't raising these questions early, they're not the right partner.
Noland Arbaugh on stage wasn't just inspiring; it was a benchmark. The devices defining the next decade of MedTech are more complex, more regulated, and more demanding to build than anything before. The old model of sequential development and late-stage manufacturing handoff won't get them across the line.
The bar is rising. The window for shortcuts is closing. The leaders who recognize that now, and choose their partners, accordingly, will be on the right side of where this industry is heading.