Latest / Ctrl AI Profit / Ep. 147 | Amazon Just Offered to Move In
Transcript
- 0:20SPEAKER_00: Frank, I want to talk about something Amazon just announced that I think is going to change the conversation around AI adoption, especially for small and mid-sized businesses.
- 0:31SPEAKER_00: What did they do?
- 0:32SPEAKER_00: They announced a brand new division inside AWS, a billion dollars committed, where they're sending what they call forward-deployed engineers, people who don't just sell you cloud services and wish you luck.
- 0:47SPEAKER_00: They embed inside your company and build AI solutions with you.
- 0:52SPEAKER_01: This is real and it's significant.
- 0:55SPEAKER_01: AWS confirmed this publicly.
- 0:57SPEAKER_01: The pitch is essentially we will put our engineers at your table inside your business to accelerate how you adopt and deploy AI.
- 1:04SPEAKER_01: It's a consulting model, but it's Amazon doing the consulting.
- 1:08SPEAKER_00: And I've got mixed feelings about it.
- 1:10SPEAKER_00: On one hand, great.
- 1:12SPEAKER_00: AI adoption is hard.
- 1:14SPEAKER_00: Most businesses are drowning in options and not sure where to start.
- 1:17SPEAKER_00: Having actual experts show up and help you build is valuable.
- 1:22SPEAKER_01: On the other hand, those experts work for Amazon, and Amazon has a very specific set of tools they want you using.
- 1:29SPEAKER_01: That's the thing.
- 1:30SPEAKER_00: This isn't a neutral consultant.
- 1:32SPEAKER_00: This is Amazon saying, let us help you build your AI stack.
- 1:35SPEAKER_00: And by the way, we'll help you build it on AWS.
- 1:39SPEAKER_01: Right.
- 1:40SPEAKER_01: Forward Deployed Engineers is a model that got famous at Palantir, which builds AI and data systems for governments and large enterprises.
- 1:49SPEAKER_01: The idea is that you don't just ship software, you embed people inside the customer so deep that they become indispensable.
- 1:55SPEAKER_01: Switching costs go through the roof.
- 1:58SPEAKER_00: It works.
- 1:59SPEAKER_00: And it works for a reason, because it actually solves real problems.
- 2:03SPEAKER_00: If you're a mid-sized business trying to build an AI-powered customer intake system or automate your back office, having an AWS engineer sitting with your team is genuinely useful.
- 2:15SPEAKER_01: There's also a talent angle here.
- 2:18SPEAKER_01: Most businesses, even good-sized ones, don't have AI engineers on staff.
- 2:23SPEAKER_01: They have IT people, maybe a developer or two, but an AI engineer who knows bedrock, knows the data pipeline, knows how to wire your existing systems into a model, that's a very different skill set.
- 2:36SPEAKER_00: And it's expensive to hire.
- 2:38SPEAKER_00: The salary for a solid AI engineer right now is not small business territory.
- 2:45SPEAKER_01: So Amazon is essentially offering to rent you that expertise.
- 2:49SPEAKER_01: You don't have to hire it, you don't have to train it, you don't have to retain it.
- 2:53SPEAKER_01: You just have to be a committed enough AWS customer to qualify for the program.
- 2:59SPEAKER_01: But let's talk about who this is actually for.
- 3:24SPEAKER_01: Exactly.
- 3:25SPEAKER_01: The value is real.
- 3:26SPEAKER_01: The strategic implications are also real.
- 3:29SPEAKER_01: You have to hold both of those things at once.
- 3:32SPEAKER_00: Right.
- 3:33SPEAKER_00: But here's why it matters for small businesses anyway.
- 3:36SPEAKER_00: Because this changes how the whole market works.
- 3:38SPEAKER_00: When AWS starts embedding engineers at scale, Google and Microsoft are going to respond.
- 3:44SPEAKER_00: And the wave eventually hits smaller businesses through managed service providers, through consultants who are trained and certified by these platforms.
- 3:55SPEAKER_01: That's exactly the pattern.
- 3:56SPEAKER_01: AWS does something at the enterprise level, and within two or three years, there's a version of it available to businesses of any size through the partner ecosystem.
- 4:05SPEAKER_01: Think about how AWS has managed AI services.
- 4:08SPEAKER_01: SageMaker, Bedrock, started as enterprise grade tools, and are now something any developer can spin up.
- 4:15SPEAKER_00: So for the small business owner listening, what does this mean for you today?
- 4:20SPEAKER_01: A few things.
- 4:21SPEAKER_01: First, this validates something we've been saying on this show that AI adoption isn't just a technology problem, it's a change management problem.
- 4:30SPEAKER_01: The fact that Amazon is spending a billion dollars to put humans inside companies to help with AI means even they know that models alone aren't enough.
- 4:39SPEAKER_01: People need guidance.
- 4:40SPEAKER_00: That's a shift from the just use the API era.
- 4:44SPEAKER_00: We're in a more mature phase now where deployment and integration are the hard part.
- 4:49SPEAKER_01: Second, if you're evaluating AI tools or cloud platforms right now, ask your vendors what their implementation support looks like.
- 4:57SPEAKER_01: Not just the documentation, not just the onboarding video.
- 5:02SPEAKER_01: What kind of human guidance comes with this?
- 5:04SPEAKER_01: Because AWS is signaling that hands-on help is a competitive differentiator.
- 5:09SPEAKER_00: I've seen this play out in my own business and with clients.
- 5:14SPEAKER_00: The businesses that actually get ROI from AI are the ones that had someone walk them through the setup.
- 5:20SPEAKER_00: The ones who just bought a license and tried to figure it out on their own are still stuck.
- 5:26SPEAKER_01: Third, and this is the strategic angle.
- 5:29SPEAKER_01: Think carefully about how deep you let any single vendor embed into your operations.
- 5:34SPEAKER_01: The value of having AWS engineers build your AI stack is real.
- 5:38SPEAKER_01: So is the lock-in.
- 5:40SPEAKER_00: And that's not a deal breaker.
- 5:43SPEAKER_00: Lock-in isn't always bad.
- 5:45SPEAKER_00: If the platform is solid, the support is great, and the cost makes sense.
- 5:49SPEAKER_00: Sometimes you want to go deep with one vendor, but it should be a conscious choice, not something you drift into.
- 5:55SPEAKER_01: The smart move is to get clear on your core workflows, the three to five processes that, if AI could improve them, would make the biggest difference to your business.
- 6:04SPEAKER_01: Then evaluate who has the best tools for those specific things.
- 6:08SPEAKER_01: Don't let a vendor's embed program drive your architecture.
- 6:12SPEAKER_00: Drive your own.
- 7:16SPEAKER_00: That's exactly what happened with the original cloud migration wave.
- 7:20SPEAKER_00: A lot of companies just let AWS or Azure's sales team design their infrastructure.
- 7:27SPEAKER_00: And 10 years later, they're paying massive bills for architecture that was optimized for someone else's revenue model, not theirs.
- 7:35SPEAKER_01: History rhyming again.
- 7:37SPEAKER_01: That's the real story here.
- 7:39SPEAKER_01: AI was never just about the model.
- 7:42SPEAKER_01: It was always going to be about who controls the layer between the model and your business.
- 7:46SPEAKER_01: And right now, all the big players are making very aggressive moves to own that layer.
- 7:52SPEAKER_00: Let me ask you something.
- 7:53SPEAKER_00: If you're a small business owner and an AWS rep calls and says, we want to send an engineer to work with your team, what's the right response?
- 8:01SPEAKER_01: Ask a lot of questions before you say yes.
- 8:04SPEAKER_01: What does this cost?
- 8:05SPEAKER_01: And does the cost change over time?
- 8:07SPEAKER_01: What happens to the systems they build if you ever leave AWS?
- 8:10SPEAKER_01: Do you own the code, the architecture, the documentation?
- 8:14SPEAKER_01: And critically, are they helping you build something portable, or are they using proprietary AWS services that only run on AWS?
- 8:22SPEAKER_00: Those are the right questions.
- 8:24SPEAKER_00: And if they can't answer them clearly, that's your answer.
- 8:27SPEAKER_01: The program will evolve too.
- 8:29SPEAKER_01: Right now it's enterprise focused, but Amazon has a track record of moving down market aggressively.
- 8:35SPEAKER_01: AWS Activate for Startups, their SMB programs, they will find ways to bring some version of this to smaller customers.
- 8:42SPEAKER_01: Here's my takeaway.
- 8:44SPEAKER_00: If you're a business owner and you haven't had a serious conversation about your AI infrastructure, not just what tools you're using, but what your actual strategy is, this week is the week to start.
- 8:56SPEAKER_00: Because the platforms are making choices right now that will shape your options for the next five years.
- 9:02SPEAKER_01: Know your stack, know who's building it, and make sure the person driving those decisions is you.
- 9:08SPEAKER_01: This is Control AI Profit.
- 9:11SPEAKER_01: I'm Michael.
- 9:12SPEAKER_01: And I'm Frank.
- 9:13SPEAKER_01: See you tomorrow.