A series of interviews on

the mechanics of business and real-world applications of machine intelligence

CTL is a series of interviews with executives at the largest global businesses. We go behind the scenes, understand how these businesses operate, and explore the most interesting applications of AI.

We have a strict no-buzzwords, no hand-waving policy. It's about near-term tactical ROI, and how we get there.

Also posting on: LinkedInTwitter
Connecting OEMs with their customers
With Jon Cooper, Operating Partner at Atlas Innovate

Key Takeaways

Manufacturers of industrial equipment want a direct relationship with their customers to improve post-purchase customer experience and capture services revenue. Offloading monitoring data, using anomaly detection, and providing in-context maintenance information through LLM-driven agents is now possible. OEMs can leverage this new technology the build the relationships they want.

Topics Covered

  • Background on Jon & Atlas
  • Pendulum is starting to swing toward OEMs wanting to managing more of the after-market business
    • Why are OEM's sensitive to providing after-market services? Why haven't they done this?
    • Who has done this well? Are there examples of OEMs taking over more of the after-market?
    • Schneider Electric with its building managment systems is a good example; it's all about helping the end-customer "press the easy button" with the product experience being seamlessly vertically integrated
  • Interesting to think about the parallels in CCC and Solera that built such amazing businesses in consumer auto and how one might replicate that here
    • The unprecedented investment in US manufacturing capacity is worsening what is already a significant worker shortage; it makes sense that something like a CCC/Solera should exist for OEMs to make those end-employees more effective
    • The state of the art today in OEM training and issue resolution is handing over a massive user manual
    • YouTube is how you solve problems at home; factory workers are no different
    • Augmented reality didn't work in the last generation to solve these issues, I think due to user experience; LLMs let you iteratively refine your query, which I think is how people want to interact with information
  • Could be an interesting vector here by which to find long-term defensibility / avoid just being a GPT wrapper; if you are the ISVs are collecting all these prior user conversations of issue resolutions, you aggregate an interesting dataset, similar to what you'd find on a Reddit or Quora or Stack Overflow -like community
  • Not only are you helping customers be more successful with your products, but you're sending very valuable data back to manufacters about how to improve their products
  • There's clearly room to improve on the existing model where things like Augury are just capturing machine data; they have no operational/human context
  • Is this type of technology more important from a preventative perspective, or post-issue response?
  • What is the right way to distribute something like this?
    • If the end user is not getting value, the whole jig is up; that's where you need to start
    • One really interesting way to engage both manufacturers and distributers would be to let manufacturers provide the context for the bot, but for distributors to be able to jump in when end-customers need a higher level of support
  • The question is, is this something revolutionary or modestly evolutionary?
  • The amount of investment into US manufacturing these days is unprecedented; can you walk us through that?
    • We are suddenly realizing that globalization of all things isn't ideal; we need availability of things like chips and medicine
    • To put some numbers to this, we are talking about a 10x increase in investment in US manufacturing y/y in some categories
    • There is so much money being pumped into the US industrial economy; there are going to be a ton of interesting companies born out of this