Meta's 53 million dollar enterprise AI hire

0:00This episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome? That's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50-page restoration block, or finally break down that long article you've had open for weeks. Gemini and Chrome is here for it. Ready to make anything online makes sense? There's no place like Chrome. Check responses set up require compatibility and availability varies 18 plus. College football is back. So, Hilton called to me the superstition concierge to make your fan rituals a reality. Need a room to match your lucky number? We got you. Want to make sure our team doesn't wash your lucky jersey? Oh, that smells lucky. Hilton's unmatched hospitality can keep up with any superstition. Even a marching bandwicker call at 555 and 55 seconds. Hit it! When you need a team that will do whatever it takes on game day, it matters where you stay. Hilton, for this day. When you need to build up your team to handle the growing chaos at work, use Indeed Sponsored

1:02Jobs. It gives your job post the boost it needs to be seen and helps reach people with the right skills, certifications, and more. Spend less time searching and more time actually interviewing candidates who check all your boxes. Listeners of this show will get a $75 sponsor job credit at Indeed.com slash podcast. That's Indeed.com slash podcast. Terms and conditions apply. Need a hiring hero? This is a job for Indeed Sponsored Jobs. Meta has hired CJ DeSci to report directly to Mark Zuckerberg and lead a newly formed enterprise AI unit. The CEO of MongoDB. Right. And the market reaction to that single-person L-change was immediate. I mean, MongoDB stock dropped nearly 20% almost instantly on the news. You are looking at billions of dollars in company value vaporizing in the time it takes to drink a cup of coffee. It really is something. The objective of this new role for DeSci is highly specific. He is tasked with taking Meta's existing artificial intelligence capabilities, the internal models and the open source versions they release, and translating them into tangible products

2:07for developers and businesses. He is there to build a software business inside a social media company. Exactly. You kind of have to look at the pressure Meta is under right now from their investor base to understand why this is happening. They have poured an astronomical amount of capital into building out their AI infrastructure. Hundreds of thousands of highly specialized computerships. Right. And entire new data centers, massive power contracts. The people holding the purse strings on Wall Street are heavily pushing for a financial return on that spending. They need a payoff. And they see the enterprise sector selling software to other big companies as the place to get it. Which sets up the core tension we are looking at. And a company that has spent its entire existence building a business model around capturing consumer attention successfully pivot to building core operational software for the enterprise. If you're not subscribed yet, take a second and hit follow on whatever podcast app you're using. It helps us keep making this. We appreciate you being here. Because they know how to keep you scrolling through a feed.

3:07We are about to find out if they know how to run the back end inventory system for a multinational retailer. The market mechanics surrounding this higher. Tell a very specific story about how hard that pivot is going to be. Yeah. We just watched a single individual's departure. Erased billions of dollars in market capitalization from a major database company. And MongoDB isn't a small startup. It is a critical piece of infrastructure for thousands of businesses. Right. That ratio of value is a perfect illustration of the premium placed on specific human expertise right now. One executive walks out the door and the market immediately decides the company he left behind is worth billions less. It assigns a massive negative value to his absence, which kind of pushes back on the conventional wisdom regarding AI. How do you mean? Well, there is this prevailing assumption right now that we are spending trillions of dollars on artificial intelligence specifically designed to automate white collar labor. The common narrative is that human workers, especially knowledge workers and managers

4:09are becoming obsolete because the software can do the reasoning for them. Right. And the executive move demonstrates the exact opposite. High end human talent, specifically the people who know how to deploy these systems within a corporate structure is becoming exponentially more valuable, not less. This episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome, that's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50 page restoration block or finally break down that long article you've had open for weeks. Gemini and Chrome is here for it. Do you need to make anything online make sense? There's no place like Chrome. Check responses set up require compatibility and availability, very 16 plus.

5:13When you need to build up your team to handle the growing chaos at work, use Indeed sponsor jobs. It gives your job post the boost it needs to be seen and helps reach people with the right skills, certifications and more. Spend less time searching and more time actually interviewing candidates who check all your boxes. Listeners of this show will get a $75 sponsor job credit at Indeed.com slash podcast. That's Indeed.com slash podcast terms and conditions apply need a hiring hero. This is a job for Indeed sponsor jobs. The compensation package attached to this move puts a very clear price tag on that value. So Meta paid to say $53 million to make the jump. If you look at how the market reacted by erasing billions from his former employer, they aren't valuing him at $53 million. No, the market is suggesting his strategic value once deployed inside Meta's ecosystem

6:17is perceived in the billions. He is being treated like a piece of foundational infrastructure himself. It makes sense. Think about the engineers who built the first electrical grids. Producing the raw electricity is one thing. That is the artificial intelligence in this scenario. The raw computational power sitting in a data center. Right. The raw output. But the electricity is entirely useless without someone who understands the grid infrastructure. You need someone who knows how to route it safely and effectively into the factories so the machines can actually run without blowing a fuse. Desire is the grid engineer. His background fits that electrical engineer profile perfectly. Yeah. Before he was running MongoDB, he held senior operational roles at Cloudflare and Service Now. Right. And Cloudflare handles security and routing for a huge portion of the internet. And Service Now is the backbone of IT Service Management for Fortune 500 companies. So he possesses a very specific, deeply technical understanding of enterprise software, cloud infrastructure, and product operations.

7:18He knows how big, slow companies buy and use software. That background highlights something inherently irreplicable about human judgment at the highest levels of corporate strategy. Absolutely. You cannot ask a large language model to figure out how to restructure an entire go-to-market strategy for a social media company trying to sell to enterprise chief information officers. I mean, the models don't understand corporate politics. Right. They don't understand the friction of enterprise procurement and they don't have relationships with the people writing the checks. Looking at the specifics of what they are actually trying to build, Meta has introduced this new unit called the Meta Enterprise Platform. Right. They already started outlining the technology stack they planned off for the businesses. And it is a distinct shift from their consumer products. What are they calling them? The product names give us a clue about their direction. They are talking about bringing the Mews agent, the Meta Business agent, the Mews API, and Mews Code to developers and businesses. These are tools designed to be embedded directly into a company's own software.

8:21When you hear a product name like Meta Business Agent, you see the underlying strategy. Meta is attempting to take their existing ecosystem, which consists of millions of advertisers and hundreds of millions of businesses who already use their platforms for marketing and customer service and transition them into enterprise software clients. Right. They want the company that buys ads on Instagram to also buy the AI agent that manages their inventory. Exactly. Decide himself outline the logic behind this transition when he took the job. He stated that the first wave of artificial intelligence tremendously changed what individuals could accomplish on a personal level. Like people using chatbots to write emails or summarized documents. Right. But he believes the next wave will force businesses of every size to completely rebuild how they innovate, how they sell, how they serve customers, and how they run their daily operations using these AI systems natively. The broader economic indicators regarding AI adoption are already validating that assumption. Yeah, consumer behavior often predicts business behavior.

9:22Like that Adobe forecast, they forecasted a 130% jump in AI-led holiday shopping traffic. We are seeing artificial intelligence move very quickly from a novelty application where you generate a funny picture to a core driver of actual commerce where the AI is routing a transaction. Meta theoretically has a unique advantage in trying to capture that commerce-driven AI market. Right. The combination of highly advanced models that rival anything else on the market. Incredibly large-scale physical infrastructure to run them and a massive existing distribution network of businesses that already have credit cards on file with Meta. The question you have to ask is whether having a distribution network for selling digital advertising translates into having a distribution network for selling core business software. To bear a question. Selling an ad campaign to a marketing department is a fundamentally different transaction than selling mission-critical infrastructure to a chief information officer. The marketing department wants reach and engagement. The CIO wants security, compliance, and uptime.

10:25The reality of enterprise software integration is completely different from consumer software. Totally. A model can be highly impressive when a consumer uses it to write a poem or plan a vacation itinerary. And that exact say model can completely fail when placed inside a complex business environment where precision is required. The friction points for enterprise adoption are severe. A business has to ask exactly where an AI tool fits into their highly specific existing workflow. You can't just drop a chatbot into a supply chain and expect it to work. Right. They have to determine with absolute certainty what proprietary corporate data that model can access safely without violating compliance laws or leaking trade secrets to competitors. And they have to define exactly what happens when the model inevitably makes a mistake. The concept of owning the exception is critical here. Owning the exception. Yeah. When a consumer AI hallucinates and gives you a slightly wrong recipe for chocolate chip cookies, it's a minor annoyance. You just get bad cookies.

11:25Right. You just toss the batch. Exactly. But when an enterprise AI drafts a flawed legal contract or hallucinates a financial metric on a quarterly earnings report or improperly routes a physical shipment of medical supplies. Right. Real damage occurs. How when inside the business has to be legally and financially responsible for that exception? The next phase of artificial intelligence competition won't be won by the company with the flashiest demo or the highest benchmark scores on a standardized test. It will be won by the company that successfully makes these models function reliably within the messy, complex, highly regulated legacy systems that businesses already use. Yeah. It's about taking a Ferrari engine and making it work perfectly inside a 30-year-old delivery truck without blowing out the transmission. Bringing in an operator with deep enterprise software experience is a deliberate move to solve those exact integration and liability problems. Right. You don't hire the CEO of MongoDB to build a better conversational chatbot for consumers.

12:26You hire him to figure out how to integrate AI into existing corporate data pipelines, how to navigate enterprise procurement cycles, and how to build a support structure that CIOs trust. We also have to consider the speed of market disruption happening right now. The world's economic dynamics are accelerating at a pace we haven't seen before. Artificial intelligence is allowing companies that have immense capital, massive compute power, and top tier talent to enter entirely new markets much faster than traditional business cycles usually allow. You can see that acceleration in how the market reacted to the MongoDB news. Investors are now being forced to reassess not just MongoDB's future without their CEO, but the competitive position of all enterprise software companies. Right. A single personnel move shifted billions of dollars, indicating how fragile current market positions are when a tech giant decides to enter your space. The fragility of corporate advantages in this environment is a crucial point for you to consider. The competitive edges that companies rely on today, their existing customer base, their

13:29established software architecture, those are incredibly difficult to protect from digital disruption when the underlying technology is moving this fast and the companies driving it have essentially unlimited resources. Like Nvidia announced a $150 billion share buyback. $150 billion. Yeah. And alongside that, a new AI security push, which is just a staggering amount of capital. Right. To put that into perspective, they are returning more capital to shareholders than the entire market valuation of most Fortune 500 companies. The companies providing the foundational compute power for all of this are accumulating resources at a speed that allows them to alter adjacent markets like cybersecurity, basically at will. Enormous capital allows these big tech companies to completely bypass the traditional startup phase. Exactly. They don't have to spend years slowly building market share, going to industry conferences and begging for pilot programs. They can immediately threaten established enterprise players simply by redirecting their

14:29existing resources and hiring away the top talent in the industry with compensation packages that no traditional software company can match. There is a cautionary tale for Meta here, though, looking at the path Google Cloud had to take to become competitive in the enterprise space. Oh, absolutely. Meta's new enterprise ambitions look very similar to the hurdles Google faced when they decided they wanted to sell cloud infrastructure to businesses instead of just selling search ads. Google Cloud a full decade to truly hit its stride. They had brilliant engineers, arguably the best in the world and unlimited capital. But proving that you can build a trusted enterprise ecosystem and establish deep corporate partnerships requires a massive amount of time, regardless of how much money you throw at the problem. Enterprise buyers want to know you will still support the software 10 years from now. Exactly. And a company known for killing off consumer products has to work very hard to earn that trust. If you look at the current state of the Meta Enterprise platform, the details are incredibly

15:29sparse. Right now, it is very difficult for enterprise businesses to even evaluate what Meta's offering because the specific architecture and go-to-market strategies haven't been clearly defined. It's an announcement of an intention rather than a fully-baked product suite. The abrupt nature of this transition is telling, decide was actually rehearsing for MongoDB's investor day just days before leaving for Meta. Wow. This strongly suggests that Meta's enterprise strategy is not some long planned and meticulously detailed roadmap that has been in the works for years. It is being constructed in real time, likely driven by the immediate pressure from investors to show a business model for their AI spending. This episode is brought to you by Google Chrome. You think you know a browser, but Gemini and Chrome? That's new. It can help you with practically anything on the web, like restoring a vintage motorcycle from a 50-page restoration block or finally break down that long article you've had open for weeks. Gemini and Chrome is here for it. Ready to make anything online make sense?

16:29There's no place like Chrome. Check responses set up require compatibility and availability varies 18 plus. Call us footballers back. So Hilton called to me the superstition concierge to make your fan rituals a reality. Need a room to match your lucky number? We got you. Want to make sure our team doesn't wash your lucky jersey? Oho, that smells lucky. Hilton's unmatched hospitality can keep up with any superstition. Want to march your bandwink up call at 555 and 55 seconds? Hit it! When you need a team that will do whatever it takes on game day, it matters where you stay. Hilton, for this day. When you need to build up your team to handle the growing chaos at work, use indeed sponsor jobs. It gives your job post the boost it needs to be seen and helps reach people with the right skills, certifications, and more. Spend less time searching and more time actually interviewing candidates who check all your boxes. The listeners of this show will get a $75 sponsor job credit at and D dot com slash podcast. That's indeed dot com slash podcast terms and conditions apply.

17:32Need a hiring hero? This is a job for indeed sponsor jobs. Metta's going to have to learn how to court enterprise developers. They have to figure out the complex enterprise purchasing hierarchy. Right. They have to learn how to sell to the people who control the IT budgets, which is a completely different skill set that optimizing algorithms for user engagement on a social feed. You have to build sales teams, support tiers and service level agreements. David Acharya has returned as the interim CEO of MongoDB in the wake of all this. He brings a very specific perspective, having previously served in that exact role for 11 years before decided to go over. He knows the company. And more importantly, he knows the current state of the enterprise software market. His perspective is particularly interesting because of the time he spent as a partner at Sequoia Capital recently. And that vantage point, looking at a massive portfolio of startups, he observed a fundamental shift in how AI is being utilized. How so? He noted that AI is expanding its role from merely helping people manage their work,

18:34like summarizing a meeting or generating code snippets to actually performing the work itself autonomously. MongoDB's immediate response to this executive shakeup highlights that shift. They hosted a sold out event in New York City focused entirely on helping developers put AI into production. Right. The entire theme was about moving from systems that just generate answers in a chat window to systems that take action at scale within a business's operational database. If we connect this push for autonomous action to the broader economic picture, it makes perfect sense. Yeah. Consider the $60 billion tariff deal between the US and China. When international trade costs fluctuate and supply chains get squeezed by macroeconomic policy, businesses immediately look to internal automation to protect their profit margins. Because they can't control the tariffs. Exactly. But they can control their internal operating costs. That margin pressure drives the need for AI that doesn't just act as a glorified search engine, but AI that can execute tasks. If tariffs increase the cost of goods, a company needs an AI agent that can autonomously

19:35reroute shipping logistics to save money. Or dynamically adjust pricing across an entire e-commerce platform. Right. Or automatically renegotiate vendor contracts. They need the software to do the heavy lifting, not just to suggest that shipping costs are too high. There is a major bottleneck in this transition toward autonomous enterprise AI, though, and it involved the workforce expected to actually use these new tools. A recent finding show that Generation Z has a significant lack of confidence in utilizing AI workflows, which could negatively impact their integration into the modern workforce. People often assume that the youngest workers, the digital natives, are automatically the most adaptable to new technology. But if the incoming workforce lacks confidence in using these complex AI tools, perhaps because they fear making a mistake that causes a massive systemic failure, or because they don't fully understand how the model arrived at its conclusion, then enterprise platforms are going to face severe adoption hurdles. Exactly. You can build the most powerful AI agent in the world, capable of optimizing an entire global

20:36supply chain. But if the employee sitting at the desk is afraid to press the execute button, or doesn't trust the output the agent provided, the software is useless. The human element becomes the limiting factor in the technology's effectiveness. This connects directly back to meta-strategy, and why hiring someone like DeSai is so crucial. Metakin build the most advanced enterprise agents available, utilizing all their compute power and engineering talent, but they still have to design interfaces and workflows that ordinary workers trust and feel confident relying on for their daily tasks. The software has to be intelligible to the person using it. We can draw a parallel to consumer behavior regarding trust and massive platforms. YouTube and meta had to backpedal and allow certain politically adjacent ads to run after initially restricting them. This demonstrates that even massive platforms with the best engineers have to constantly adjust their algorithms and their policies based on unpredictable human reactions, cultural shifts, and market realities. You can't just code your way around human behavior.

21:39The ultimate test for the meta-enterprise platform won't happen in the boardroom when the CIO signs the contract. It will happen at the desk level, months later when thousands of employees are actually asked to use it. The success of this initiative relies entirely on user trust and adoption on the ground floor. If the employees actively resist the integration, or find workarounds because they don't trust the AI, the entire enterprise deployment fails, regardless of how good the underlying model is. We're watching a social media giant attempt to buy its way into the highly complex, highly regulated enterprise software market by acquiring the exact human talent needed to navigate corporate bureaucracy. They are betting that their consumer infrastructure can be retrofitted for corporate operations. Even with the best talent in the world designing the strategy, the real question is whether traditional conservative businesses will ever feel truly comfortable handing their most sensitive, mission-critical operational data over to the company that built Facebook.

22:44We appreciate you being here. Also check out our YouTube channel for more business and tech updates. There's a link in the description.