The hardest part of physics AI is not building the models. It is getting them to travel the last mile inside a real engineering organization, where workflows have been tuned for decades, certification regimes set the bar, and decisions are made by people who have been right for thirty years. The platforms exist. The models work. Customers are paying attention. What remains is the work of taking the technology into industries that will not adopt it on the strength of a benchmark, and earning the right to change how those industries design, manufacture, and operate the things they make.
That work is done by people, not by software. Luminary is built around two roles that we believe are the most consequential in the company, the Industry Practice Lead and the Lead Delivery Engineer, and we have organized the company so that the people in those seats are set up to do the best work of their careers.
Industry Practice Leads
The IPL owns the customer side of the bet, which means walking into a CEO’s office and prescribing a transformation rather than a tool, scoping the engagement, structuring the program, and staying accountable for the outcome. They know the workflow, the external regulator, the sign-off, and the political weather inside the customer because they have spent a career in one of the industries we work in, whether that is aerospace, automotive, defense, or electronics, long enough to be the senior person in any room.
What defines the IPL is the ability to see the transformation a customer needs before the customer can articulate it, design the path from where they are today to an engineering organization built around physics AI, and stay credible across the years it takes to get there. That is closer to the work a senior partner does inside a top consulting firm than it is to enterprise software sales, with one important difference: the deliverable is not a recommendation, it is a customer whose engineering process has actually been rebuilt. Our IPLs are vertical because the credibility required to lead that transformation requires deep industry expertise.
Lead Delivery Engineers
The LDE owns the technical side of our engagements and runs a matrix team that turns a model into a decision an engineer can defend. That means scoping the data, building the model, integrating it into the customer’s tooling, and shipping something that actually runs on a Friday afternoon when an engineer needs it, with the LDE as the technical owner of the engagement from day one to last day.
The LDE is not a consultant. They are a senior engineer who has shipped real things, can hold their own with a chief engineer, and can lead a cross-functional team of physicists, ML researchers, and platform engineers on the strength of being technically right rather than on a reporting line.
Why Both, and Why Together
A great IPL without a great LDE writes a check the team cannot cash, and a great LDE without a great IPL builds something impressive that no customer adopts. The seam between them is where customer transformation actually happens, and Luminary is organized around making that seam strong.
Physics AI is going to reshape how the physical economy designs, manufactures, and operates the things it makes, on a timeline shorter than most people in industry are willing to admit. The companies that figure out how to put physics intelligence inside their engineering process early are going to design better products, qualify them faster, manufacture them cheaper, and operate them with margins the rest of the industry will not be able to match. The companies that wait will spend the next decade trying to catch up to a cost structure and a design capability that has already left them behind. The people who lead that transition inside their industries, on the IPL and LDE side of the table, are going to define how an entire generation of physical products gets built, and there is a narrow window in which that work is still being shaped rather than executed. If you have spent a career making real things real, this is the move worth making, and it is worth making now.
