This interview is about MongoDB CEO Dev Ittycheria's view on the database opportunity in the AI era. He compares AI's impact to the iPhone App Store evolving from silly apps to Uber—AI today is basic chatbots, but real transformative apps are coming, and MongoDB could be key infrastructure. He's optimistic but says it's still early. Key holdings: MongoDB (revenue grew from ~$30M to ~$2B, cloud share from 2% to 68%), Datadog (early investment with 5% monthly growth, now >$2B revenue), NVIDIA (high valuation but real demand; no position stated).
MongoDB CEO Dev Ittycheria discussed in an interview the evolution of the database industry in the AI era, emphasizing the company's journey from a startup to a developer data platform serving tens of thousands of customers globally (covering 100 countries). The core view is that leaders need a high
Dev Ittycheria has served as MongoDB's CEO since 2014, leading the company from roughly $30 million in revenue to nearly $2 billion in annualized revenue, and taking it public in 2017. This interview's main thread explores his leadership philosophy as a technology company CEO, his approach to building an enterprise sales organization, and how MongoDB positions itself in the AI era. The most impactful takeaway: AI's impact on databases will mirror the evolution of Apple's iPhone App Store — moving from simple "flashlight" level applications to deeply transformative ones like Uber and Airbnb, with MongoDB poised to become core infrastructure in this wave.
Dev Ittycheria uses the evolution of the iPhone App Store as an analogy for AI's impact on the database industry. In the first year after the iPhone launch, the App Store only had "trivial" applications like "flashlight" and "iBeer" (a beer-drinking simulation). But once people truly embraced the iPhone as a serious computing device, Uber, Airbnb, and enterprise iPhone applications were born.
"The same thing will happen in the AI world — we will move from simple chatbots that answer questions to something far more profound that uses real-time data to make intelligent business decisions."
Today, the value of AI "first accrues to the bottom of the tech stack" (Dev bluntly notes "NVIDIA isn't trading at 300x P/E"), but ultimately the returns will materialize through the application layer — improving customer experience, automating cost reduction, creating new business models, or disrupting existing ones.
The signal MongoDB observes: "thousands of startups are already building AI applications on MongoDB," which is seen as a "crystal ball for where large enterprises are heading." Yet Dev emphasizes "it is still very early days."
The key debate centers on AI's impact on database architecture. Dev points out that the switching cost for large language models (LLMs) is currently "actually very low" — it's easy to switch from OpenAI to Anthropic or Llama. But if LLMs start embedding memory, storing user history, "all that historical data lives in a persistent data storage layer, making switching costs much higher." This means databases will play an "increasingly important role" in the AI era.
Dev Ittycheria's core leadership belief: "To drive excellence, you must be extremely picky, and most people shrink under pressure, afraid to make the decisions required for true success."
He observes a common problem: people "don't dare hold accountable those who need to be held accountable," "don't dare be extreme in product positioning or pricing decisions," and "don't dare be extreme in go-to-market strategy." The result is a large number of "very mediocre companies."
He particularly highlights the harm of passive-aggressive behavior: "I see passive-aggressiveness as a form of duplicity. You must create a culture where people can freely and constructively express their true views on an idea or decision." One of MongoDB's core values is "intellectual honesty."
"If you see a problem but take no action, that problem is no longer that person's problem — it's your problem. If you have the power to fix it but don't, then you are the problem."
Dev believes the common fallacy in B2B is over-focusing on the product ("if you build a great product, they will come"), but the reality is "the best technology doesn't always win." He recalls early experiences at IBM, HP, and other large vendors where "account control" made customers choose "safe decisions."
"Magic happens when you combine a great product with a great go-to-market organization."
Hallmarks of a great enterprise sales organization:
1. Broad-based performance: Avoid the "80-20 rule" (20% of sales reps contribute 80% of results); create an environment where everyone can perform well.
2. Hire hungry, smart people who want to make a name for themselves.
3. Develop talent: Not just product skills, but how to progress deals, evaluate opportunities, and forecast — Dev stresses that "in a high-growth business, revenue forecast/booking forecast is a proxy for your expense forecast; bad forecasting burns cash quickly."
4. Execution consistency: Not a "yo-yo" — one good quarter, one bad — but consistent delivery.
For long sales cycles, Dev breaks them into three core questions: "Why does the customer want to do anything (pain point) → Why is MongoDB the best choice → Why now." He warns: "You should never respond to an RFP you didn't participate in writing, because it was definitely written by another vendor."
"The common mistake most people make when hiring is focusing only on background, experience, and skills. You need to understand the person's psychology — what drives them, what motivates them, who they want to become."
Dev particularly values the "Jeremy Giffon assertion" — "the only sustainable advantage in life is a psychological advantage, because human nature doesn't change." He looks for people who can manage their ego, accept low status, delay gratification, and have a long-term orientation.
A common interview question he uses: "What is the hardest thing you have ever been through?" He shares his own story — coming from a broken family, being an Indian Christian (a minority), immigrating across multiple countries due to the stigma of divorce, and always wrestling with "the question of 'Am I good enough?'" He believes such experiences shape "inner drive" and that "you can't necessarily escape your upbringing."
Dev's investment framework has three elements:
1. Large market: "Your outcomes are directly correlated with market size."
2. Defensible technological advantage: Evidence the company has a durable edge.
3. Smart and coachable CEO: "There is no compression algorithm for experience. Some people just learn faster than others."
He shares a counterintuitive observation: "There is an inverse relationship between effort and outcome — your best portfolio companies, you show up, eat a donut, and leave, because they're doing so well; the companies where you spend the most energy helping the management team rebuild are likely the ones struggling."
Signals of product-market fit: "Do customers come back and buy more? Anyone can buy something once as a science experiment. The key is whether you can repeatedly get more business from the same customer, or show a pattern of winning other customers with the same use case." He specifically mentions the Datadog investment example — early on, "5% month-over-month growth," indicating "not only was selling easy, but customers grew very quickly after purchase."
The most useful framework Dev learned from Andy Grove: "People fail for only two reasons — either they lack the skill to do the job, or they lack the will (drive) to do it." He calls it the "skill-will matrix."
His three-step accountability process:
1. Set expectations very clearly — "Vague expectations set people up for failure because they don't know what good looks like."
2. When expectations are not met, first blame yourself — "I may not have explained clearly enough why this expectation is important."
3. If the expectation is missed again, accountability becomes easy — "Because I've already done steps one and two."
"Most people skip steps one and two, so at year-end you give a 'meets expectations' rating, and the person says 'I thought I was doing great,' and the conversation becomes very dysfunctional."
Dev's biggest concern: "If people feel MongoDB is not the right architecture for the best AI applications," or "if the next generation of companies and applications feel MongoDB is not for them." He admits: "It's easy to become complacent."
Regarding MongoDB's business characteristics, he acknowledges: "We are not like HR platforms where everyone standardizes on Workday, nor CRM where everyone standardizes on Salesforce, nor even data warehouses — we have to win application by application, workload by workload." This makes sales costs relatively high, and reducing customer acquisition cost while creating viral "developer self-selection" remains a persistent challenge.
On whether AI is overhyped, Dev believes "people always overestimate the short-term impact of new technology and underestimate the long-term impact," because technology adoption follows an S-curve. He believes AI will be "extremely transformative," but its delivery will come through the application layer — just as software democratization enabled small and medium businesses to use the best CRM/HR software, AI will enable them to access the best legal, accounting, and financial services, "ultimately a net positive for everyone."
| Ticker | Analyst Stance | Key Data |
|---|---|---|
| MongoDB | Bullish (CEO himself) | 2014 revenue ~$30M → approaching $2B annualized revenue; cloud business from 2% of revenue in 2017 → 68% in 2024; was still loss-making at IPO in 2017 (-38% operating margin) → 16% operating margin in 2023 |
| Datadog | Bullish (early investor) | Revenue ~$1M at time of investment, now over $2B; early month-over-month growth of 5% |
| AppDynamics | Bullish (portfolio company, mentions CEO Jyoti) | Acquired by Cisco; CEO Jyoti described as "extremely smart and coachable" |
| NVIDIA | Neutral (valuation comment) | "Not trading at 300x P/E"; demand is real, but Dev did not explicitly state a position direction |
| Oracle | Risk (as a cautionary case) | "Oracle is not the most beloved vendor, but there are still a massive number of Oracle databases because migration costs are extremely high"; "I can't think of a single startup today that is building on Oracle" |
| Snowflake | Neutral (as a comparison reference) | Mentioned as a competitor in the "data warehouse/standardization platform" category |
| PTC | Positive (company Steve Walski previously led) | 40% operating margin in the 1990s, 100%+ growth for 10 consecutive years |
1. Dev Ittycheria: "AI applications are replaying the story of the iPhone App Store — from flashlight to Uber, just 100x larger." Current simple chatbots ≈ the iBeer of its day; truly transformative applications have yet to arrive. MongoDB observes the technology choices of thousands of AI startups to anticipate enterprise direction.
2. Dev Ittycheria: "To drive excellence, you have to be extremely picky. Most people cave under pressure and don't make the decisions needed to be truly successful." He equates "not facing the problem" with "accepting mediocrity," and it punishes all the hardworking good people.
3. Dev Ittycheria: "If you see a problem and don't act, that problem is no longer that person's problem — it's your problem." This is his action principle: having the power to solve a problem but failing to act makes you the problem itself.
4. Dev Ittycheria: Andy Grove's "Skill-Will Matrix" — People fail for only two reasons: lack of skill or lack of will. Using this 2×2 matrix to evaluate a team quickly identifies the root cause of issues.
5. Dev Ittycheria: Three-step accountability method — first, clearly define expectations; second, on the first mistake, take ownership yourself (communication wasn't clear enough); third time, hold real accountability. Most managers skip the first two steps, resulting in year-end reviews turning into dysfunctional conversations of "I thought I was doing great."
6. Dev Ittycheria: "People perform to the level they are inspected, not the level they are expected." If trusting adults to do their work autonomously were enough, fewer managers would be needed. High inspection creates a transparent culture with "no place to hide" and also fosters "action orientation."
7. Dev Ittycheria: "A players hire A players, B players hire C players, C players hire F players." The most reliable signal of whether a leader is growing is the quality of their team — if top talent starts leaving, the problem must be with the leader. (Readers should note this is from a position-holder's perspective; such aphorisms carry narrative elements.)
8. Dev Ittycheria: "The sign of product-market fit is that customers come back to buy more." Any customer can buy something as a "science experiment." The key is whether you can repeatedly get more business from the same customer, or demonstrate the same use case to acquire other customers. (NDR > 120% is a good signal in the public company version.)