About
I specialize in walking into organizations that are stuck and getting them moving again.…
Services
Articles by Brian
-
Part III Market Management
Part III Market Management
Market Management The product manager should know the global market for their product in more detail than anyone else…
2 Comments -
Management of Physical Products Part II Task-Time Management
Management of Physical Products Part II Task-Time Management
Today I will be continuing my series on product management of physical products reviewing the task-time management…
-
Keys to Successful Product Management of Physical Products
Keys to Successful Product Management of Physical Products
I will be writing a 3 part series regarding what I believe are some critical keys to successful product management…
Activity
1K followers
-
Brian Sandberg shared thisI'm excited to share that I've officially earned my ASQ Certified Lean Six Sigma Green Belt certification! This certification has strengthened my understanding of process improvement, statistical analysis, and continuous improvement methodologies, and I'm looking forward to applying these tools to help organizations improve quality, efficiency, and operational performance. Learning never stops. #LeanSixSigma #ASQ #ContinuousImprovement #OperationalExcellence #Quality #LeadershipASQ Certified Six Sigma Green Belt • Brian Sandberg • ASQ CertificationASQ Certified Six Sigma Green Belt • Brian Sandberg • ASQ Certification
-
Brian Sandberg posted thisPart 5 - The Next Constraint: What Problems Are Worth Solving If AI creates excess capacity, the question isn’t what jobs go away—it’s what problems we finally have the ability to solve. Throughout this series, I’ve argued that AI is shifting the constraint—from labor to observed outcomes from real systems. But if that’s true, then there’s a second-order effect: What do we do with the capacity that gets freed up? Historically, most organizations have been forced to focus on short-term, ROI-driven work. Not because it was optimal—but because it was necessary. If that constraint starts to loosen, we have the opportunity to take on a different class of problems. The kind that are: Long-horizon Experimentation-heavy High uncertainty Difficult to justify under normal constraints A few that stand out: Abundant, low-cost energy — driving the cost of energy down to the point where it is no longer a limiting factor Materials science breakthroughs — new materials that unlock advances across manufacturing, infrastructure, and transportation Space-based infrastructure — expanding manufacturing and resource access beyond Earth Human longevity — extending healthspan through sustained biological research and experimentation These aren’t new ideas. What’s new is the potential capacity to pursue them at scale. Not as side projects—but as coordinated, sustained efforts. This is where large organizations—and in many cases, government-backed or mission-driven groups—will likely lead. Because these problems require: Time Resources Coordination And a tolerance for uncertainty AI doesn’t just improve efficiency. It changes what we can afford to work on. And over time, that may be the most important shift of all.
-
Brian Sandberg posted thisPart 4 — The Larger Shift: What Do We Do With the Capacity? If AI continues to remove constraints, we will eventually create excess capacity across the system. The question is: where does that capacity go? Historically, organizations have been constrained by time, labor, and cost. That forced a focus on short-term, ROI-driven work. But if AI reduces those constraints, something changes. We gain the ability to redirect capacity into areas that were previously out of reach. Long-horizon problems. High uncertainty work. Efforts that don’t produce immediate returns. There’s a historical parallel here. Think about the scale and focus behind the Manhattan Project at Oak Ridge. It wasn’t driven by short-term ROI. It was a coordinated effort—massive resources, aligned talent, and a singular objective—focused on solving a problem that required years of sustained effort and experimentation. That kind of work is rare because it’s expensive, uncertain, and hard to justify under normal constraints. AI changes that equation. If productivity continues to increase, we create the opportunity to take on more of these types of efforts—projects that require scale, patience, and deep experimentation. This likely won’t happen evenly. It will be led by large organizations—and in many cases, government-backed or mission-driven efforts—because they’re the only ones with the scale and time horizon to support it. AI doesn’t just improve efficiency. It expands what we can afford to work on. And over time, that may be the most important shift of all.
-
Brian Sandberg posted thisPart 3 — Competitive Advantage: Who Owns the Data If the new constraint is observed outcomes from real systems, then the next question is: Who is going to own it? At a basic level, small companies will use AI to become more efficient. They’ll reduce costs, automate tasks, and in some cases gain market share. If they are a leading small organization, they will invest in creating offline data that is proprietary to their organization. But I think most will be happy with the incremental increase that normal AI use is going to provide. But large organizations have a different option. Instead of reducing headcount, they can repurpose people toward generating new, high-value data: Running experiments Capturing real-world outcomes Documenting edge cases Feeding systems with grounded truth Over time, that data compounds. And that’s where the moat gets built. Not just better models, but better data that no one else has. There’s a parallel here with private equity. Firms are moving down market, acquiring smaller companies, and building scale through operational leverage. The best operators don’t just cut costs—they reinvest in capabilities that compound over time. AI creates a similar opportunity. The organizations that win won’t just use AI to reduce labor. They’ll use it to free up capacity—and redeploy that capacity into areas that weren’t feasible before. If that happens, it becomes increasingly difficult for smaller companies to compete—not because of capital alone, but because they don’t have access to the same depth of real-world learning. In manufacturing terms, AI removed one constraint. Now the question is: Who is going to own the next one?
-
Brian Sandberg posted thisPart 2 — The New Constraint: Observed Outcomes from Real Systems In Part 1, I talked about the risk of convergence if AI continues to train on the same data. So what’s the constraint? From a systems standpoint, AI is removing one major limitation: human labor on existing tasks. That’s increasing speed and efficiency across the board. But every time you remove a constraint, another one takes its place. So what’s the new constraint? Not labor. Not compute. Observed outcomes from real systems The bottleneck becomes high-quality, real-world data that can’t be simulated. AI can generate content. It can simulate scenarios. But it can’t fully replace observed outcomes from real systems. And that matters. With the same number of people interacting directly with real systems means the same amount of : Experiments Observations Failures Edge cases Which are exactly the inputs that create new learning. If convergence is the risk, then observed outcomes is the counterbalance. So the real question becomes: How do we increase the rate at which we learn from real systems? We must expand the number of people creating new data at a rate as fast as we are saving on resources that AI is able to efficiently perform. That feels like the next constraint to solve.
-
Brian Sandberg posted thisIf AI is trained on the same data, do all ideas eventually start to look the same? This is Part 1 of a 5-part series on my perspective of AI through the lens of constraint thinking, where we are today, and where this may lead. I think we’re heading toward a convergence problem. Most AI systems are trained on large, overlapping datasets. As adoption grows, more content is generated by AI and fed back into those same systems. Over time, you start to see the effect: The outputs begin to converge. The ideas begin to converge. The solutions begin to converge. We get more of what already exists: just faster. Less novelty. Less differentiation. That’s not immediately obvious because productivity is going up. But underneath that, there’s a question worth asking: If we keep training on the same data—and increasingly on AI-generated versions of it: Where does new insight come from? Curious how others are thinking about this.
-
Brian Sandberg shared thisGamma.app is an impressive tool worth exploring if you’re not already using it. I’ve worked with other presentation platforms, but none have felt this seamless. It accelerates the packaging of ideas while still giving you full control of the message. AI handles the structure. You control the substance.
-
Brian Sandberg shared thisI love manufacturing, fishing, and motorcycles, probably for the same reason. They reward patience, systems thinking, and discipline. ChatGPT did a nice job capturing that visually.
-
Brian Sandberg shared thisLocal optimization doesn't fix the system. On-time delivery isn't a machine problem; it's a system problem. We once turned around poor OTD without touching a single manufacturing process. Instead of chasing utilization, we challenged the assumed lead time with the 5 Whys. By the third "why," the policy constraint crumbled. We talked directly to customers; key insight: Several major clients were stocking product and didn't need our standard lead time. We realigned to actual demand: Pulled real forecasts Sized batches to protect flow (not individual machine efficiency) Cut WIP and excess inventory Scheduled around the true system constraint Results: Record-high on-time delivery No shop floor changes required Stronger, trust-based relationships with top customers The Goal takeaway: Optimize local efficiencies all you want, if you're not subordinating everything to system throughput driven by real customer demand, you're optimizing the wrong thing. What's a "sacred" lead time or assumption in your operation that might actually be the constraint? Share below. #TheoryOfConstraints #Manufacturing #OnTimeDelivery #ProcessImprovement #Lea
-
Brian Sandberg liked thisBrian Sandberg liked thisTwo weeks into being laid off, I’ve learned something surprising. Apparently, layoffs have become a marketing opportunity. My inbox is full of messages from “recruiters” offering exciting opportunities, only to find out they’re selling resume services, coaching programs, job search subscriptions, or other paid products. People who have been laid off are often dealing with financial stress, uncertainty, and a loss of confidence. They deserve support—not sales pitches disguised as opportunities. A little empathy goes a long way. Not every vulnerable moment needs to become someone’s business opportunity.
-
Brian Sandberg liked thisBrian Sandberg liked thisFor well over a century, Honeywell has helped solve some of the world’s toughest, mission-critical challenges. On June 29, we’ll launch the two companies that will carry that remarkable legacy forward. Meet the new: Honeywell Technologies will be a pure-play automation company, leading the industrial sector's transition from automation to autonomy by pairing our deep domain expertise with decades of data from our global installed base. Honeywell Aerospace will build on its leadership across commercial aviation, defense and the future of flight—bringing focused innovation to the technologies and systems that keep the world connected. This is the start of an exciting new era for both businesses! Our announcement: https://hwll.co/az4lt0j3 My interview with the Financial Times: https://hwll.co/zqvw885w
-
Brian Sandberg liked thisBrian Sandberg liked thisIt’s a bad time to enter the trades! Manufacturing hiring is down 40% and the trades sector lost 150,000 jobs in the last year. This won't surprise anyone running a trade school. We've been watching it in our own data for a while. Tariffs and high interest rates are making it hard for businesses in the trades to grow. Homeowners are delaying projects, commercial builds are getting completely shelved. The tradespeople who can are good at talking to clients will make it... But if you’re a high school or college student - consider going into healthcare and education instead.
-
Brian Sandberg reacted on thisBrian Sandberg reacted on thisA little Bob Seger today; Turn the Page! 🎶 Excited to officially announce my new role with U.S. Bank as Vice President | SBA Business Development Officer. #bbf #sba #smbfinance #businessacquisitions
-
Brian Sandberg reacted on thisBrian Sandberg reacted on thisAfter nearly 3 years with my current lender, I’m grateful 🙏🏻 for the relationships, opportunities, and referral partners I’ve had the chance to work with and become friends with along the way. This will continue. #bbf Over the past 12 years in SBA lending, I’ve primarily focused on business acquisition financing in the $1.5MM-and-below range. Excited for that scope to expand!!🙌 Now able to pursue: • SBA 7(a) and 504 opportunities utilizing the full limits of both programs 💰💰💰 • Ground-up construction • Pari Passu structures • Larger cash-flowing acquisitions that make sense • And yes… the TRADES ✅ Let’s just say the next chapter involves one of the top 5 banks in the country and a team that originated nearly $1 billion in SBA lending. 🚀 Official announcement coming soon. 👀
Experience
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Education
-
Texas McCombs School of Business
-
-
Graduate Program with 6 modules, Python Foundations, Advanced Machine Learning, Introduction to Neural Networks, Natural Language Processing with AI, Introduction to Computer Vision, Model Deployment
-
-
-
-
-
-
-
Licenses & Certifications
Languages
-
Spanish
Limited working proficiency
Recommendations received
1 person has recommended Brian
Join now to viewOther similar profiles
Explore more posts
-
Lorena Acosta
Most prototype delays don’t start in engineering. They start in assumptions. Assumptions about how a part will be machined. Assumptions about material availability. Assumptions about tolerances that “should be fine.” And those assumptions don’t usually show up until the part hits the shop floor. After 20 years working alongside engineers, machinists, and production teams, I’ve learned something: A prototype drawing might look perfect in CAD… …but manufacturing always has the final vote. I’ve seen things like: • Parts designed with internal features that couldn’t physically be machined • Tolerances tighter than the machines could realistically hold • Materials that looked great on paper but added weeks of lead time • Designs that required three setups when one would have worked None of these were bad engineering. They were just missing manufacturing context early in the process. The best prototype programs I’ve seen all do one thing well: They bring manufacturing into the conversation earlier than expected. That’s often the difference between: 1 prototype iteration and 4 redesign cycles. For engineers and product teams here: What’s one small design detail that caused a surprisingly big prototype issue? #Engineering #Manufacturing #Prototyping #ProductDevelopment #R&D
-
Dustin Smith
Getting your whole team aligned on GD&T fundamentals is one of the smartest investments a company can make. When everyone—from machinists to programmers to inspectors—shares the same understanding of how parts are defined and measured, everything flows smoother: ✅ Less time spent debating print interpretations ✅ Fewer nonconformances and rework ✅ Clearer communication between engineering and the floor ✅ Consistency in how datums and tolerances are applied and verified At Assurance Technologies, we love seeing teams take that next step toward speaking a common language of quality. Training like this builds confidence, improves collaboration, and ultimately strengthens your manufacturing process from design to inspection. If you’d like to bring GD&T training to your facility or join one of our regional sessions, reach out—we’d be happy to help your team get on the same page. https://lnkd.in/gMYx5EqC #GDandT #Metrology #Quality #Manufacturing #Training #AssuranceTechnologies
2 Comments -
Robert Fey
Corrosion = a cost running into billions. ⚠️ In this blog, Corrdesa LLC and Simcenter STAR‑CCM+ help the U.S. Air Force tackle rust in critical aircraft parts, by moving from manual testing to fast, accurate virtual predictions. Read the blog to learn how simulation is changing aerospace maintenance: https://sie.ag/7FQR2Y
-
Bryan Kreager
Are you familiar with the FAA PMA process? It’s one of the most misunderstood parts of the aerospace supply chain — and it’s far more involved than many people realize. We’re starting a new series breaking down what the process actually looks like from the inside. In Part 1, we cover: 💠 What FAA PMA really requires (it’s not just manufacturing approval) 💠 Why aerospace seals are considered safety-critical components 💠 How reverse engineering works in a regulated environment 💠 Why we chose the test and computation path For teams responsible for supply chain reliability, engineering validation, or vendor qualification, PMA approval represents a lot more work behind the scenes than most people see. Curious — for those working in aerospace: What part of the PMA process do you think is the least understood?
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content