Document

Exhibit 99.2

To our shareholders:

One of the earliest debates we had at DoorDash was whether we ought to first build the consumer app or the Dasher app, the product that Dashers use to make deliveries. It seemed so intuitive that we ought to first build the consumer app because isn’t that what we are? To tilt the decision even more in favor of shipping an improved consumer app, our consumer experience at the time was a website with no photos, no search, and (very) limited functionality.

Yet, as founders fulfilling deliveries everyday at that time, we knew that bringing you a burrito - on time, in the condition you’d want, every single time - was a lot harder than it sounds. As much as we had wanted to immediately improve our consumer product, we knew that the job to be done for consumers was ultimately to deliver them their order, which meant we had to solve for the entirety of the consumer experience. At that time, building the Dasher app was the best way we could improve the value we offered to our consumers.

Systems vs Products

The physical world is complex, unpredictable, and often indeterminate, which results in challenges at every step of an order. What if the consumer forgot to update their address while ordering dinner for home while at work? What happens when a particular item is out of stock or there happens to be a long line of other tickets? What vehicle type ought to be considered and how do we match the order with a Dasher’s preferences? Where should they park and what happens if the order is getting assembled from different stations? How do you take the least trafficked route or how do you know which gate in an apartment complex is best for entry? This list of questions may sound like overkill to think about when you just want DoorDash to bring you lunch and dinner, but they represent a small fraction of the complexity we must solve in order to delight millions of consumers every day on our platform.

Before getting into all of the complexity of the systems we operate today, even in 2013, we knew that building DoorDash would require much more than just building a consumer app or single product. We had to build a collection of products that interfaced and interacted with one another to bring you that burrito, on time and with high quality, every time. Doing so reliably and efficiently requires building systems.

Building systems

Systems are orchestration of products whose job is to provide a complete and high quality experience for its customers. Today, DoorDash has built multiple systems. When you order from the DoorDash app, our Marketplace system orchestrates almost a dozen products, each with multiple components that must work together. For example, a fraud detector and a refunds and credits engine might be two components of the Support product.

Critically, components within products and products within a system must be coordinated and maintain state. Let’s suppose a consumer orders dinner and DoubleDashes grocery items to prepare lunch for the next day, and there is an item that’s out of stock for his sandwich. Then all of the components - the substitution builder from the catalog system, the refunds and credits engine, the ETA service, the promotions tool, just to name a few - need to maintain the same state in order to correctly communicate updates to the order and fulfill it as the consumer wishes.

The challenge, of course, is that each of these components has their own objective functions. Kept in isolation, managing to individual component goals could be a disaster. For instance, what good would it be if we built the best user interface for our consumer app only to show up late on every delivery? The job of each component is ultimately to help all of the audiences win.

When done right, building cohesive systems allows us to build the best customer experience at the lowest cost. To do so, we have to correctly build the vertical and horizontal integrations of these components and carefully manage the complex coordination of the handoffs between components. If we do this poorly, say we miss components or get the interaction design between them wrong, then our system would produce mistakes, which leads to worse customer retention and

higher unit costs. When compounded over time, even small gaps in the quality of our components or the coordination between them can be the difference between success and failure, both for our customers and our shareholders.

Getting it right therefore is extremely important and equally daunting. To build great systems, you have to operate at the lowest level of detail. This is why, since our founding, we’ve asked employees to WeDash, a tradition in which we fulfill deliveries ourselves to understand how our products interact and how our systems succeed or fail at solving the challenges associated with building for the physical world.

Just as important, building great systems requires an obsession over our customers. There is a proactivity associated with this lens. If you take this lens, you tend to improve your system before being asked to do so. You’ll likely audit what’s working (or not) even without a compliance request. Ultimately, it’s this proactivity and persistent desire to get a little bit better every day that will drive you to perform thousands of experiments to produce enough discoveries to remove delays, inefficiencies, and inaccuracies so that you continually create greater surplus for your customers. That’s what we’ve done at DoorDash. In each of the past five years at DoorDash, we've generated more savings for more DashPass members, driven more sales to more merchant partners, and created opportunities for more Dashers to earn more money than the year before. And we think there is more to do.

Next building blocks

In some ways, I feel like building the future of DoorDash today is very similar to how we first started building DoorDash in 2013. At our core, we’re building systems for local merchants to be successful as they evolve from physical to omnichannel businesses. The difference now is that we have more products, audiences, and geographies to consider.

If the complexity of bringing you a burrito on time is hard, think about the systems we must build today. In the past several years, we’ve grown from a single product (restaurant delivery) operating in the US into a multi-product portfolio serving over 40 countries around the world. In addition to our Marketplaces, we are now building several new systems within our Commerce Platform to help merchants be successful on their own digital channels and within their four walls. To give you a sense of this complexity, consider a few of the differences to solve for in our systems: delivering alcohol vs a 64” TV vs french fries, integrating our systems into a merchant’s systems in order to fulfill and shop for orders through their channel vs on our Marketplace, or geographic differences such as the grid-like large layouts of US cities vs some of the non-grid like denser cities in Europe. These examples represent a fraction of the cross product of challenges that reflect more than a dozen categories for deliveries, tens of thousands of cities, and multiple channels for fulfilling consumer demand. Despite this volume of new challenges, our approach to solving them is the same: build systems that stay in state in order to completely solve the jobs to be done for our audiences.

In addition to our growth and evolution, there are also new technologies to incorporate. While AI agents haven’t yet achieved their full potential, it’s clear that adding the reasoning power behind LLMs can produce more intelligent systems. But this will require careful investigation as LLMs require verifiable responses to produce great outcomes, and things can go haywire when they don’t have the right context or aren’t connected to the appropriate workflows. And some things - especially in the physical world - have no perfect answers. These cases require judgment.

The same promise and challenge can be said about autonomous vehicles (AVs). AVs by land and air will undoubtedly become a part of our systems, and we are pushing aggressively to make this happen. Some of these products will be built by partners and others will be built internally, like DoorDash Dot, our purpose built vehicle for delivery that travels the road, bike lanes, and sidewalks.

Regardless of technology, AVs, AI agents, or chatbots alone - just like a consumer app - cannot solve the entirety of a customer’s problem. These products will become part of our systems at DoorDash. Like we have done with our existing systems, in order to actually solve the customer’s problem, we must get to the lowest level of detail and get the interactions right between these products and their underlying components. That’s why for example, in the case of AVs, perhaps the most valuable part of what we’re building is our Autonomous Delivery Platform, which will match vehicles to routes and manage the handoffs between AVs and Dashers, and everything in between. Put another way, even as we incorporate new technology to improve our stack, we’re still building systems.

We currently operate our systems across three different tech platforms at DoorDash, Deliveroo and Wolt. This slows us down. Consequently, one of the systems we’re building is for ourselves as the customer, a single technology platform that will allow us to better leverage our global scale while still enabling us to offer local audiences the best possible experience. This is a massive and expensive undertaking and honestly one you shouldn’t do if you thought your best days were behind you. On the contrary, one of the lessons we’ve learned in building DoorDash is that to build systems that endure, you must also think and invest in the long term. That means sometimes you have to start over. For instance, we could have taken the DoorDash platform and subsumed Wolt and Deliveroo. We could have made our codebase less malleable to incorporate AI. These changes would have taken less time and cost a lot less to build. But that could lead to disastrous results for customers. Instead, we’re choosing to build a new platform that will incorporate the best of what we’ve learned across each of our brands all while running our existing systems that power the largest local commerce platform outside China.

Building systems that completely solve the needs of customers is at the core of what we do at DoorDash. To build them successfully, we have to operate at the lowest level of detail, obsess over customers, engage proactively, invest over the long term, and sometimes even start over. At times, this is a painful exercise, but when done right, it leads to compounding surplus for all of our audiences. It certainly helps when you have a great team like the one we have at DoorDash, a group I’m lucky to be a part of, whose standards are high and are always looking to raise them.

Thank you for investing and building alongside us.

Tony Xu

CEO and Co-founder, DoorDash

February 2026

Forward-Looking Statements

This shareholder letter contains forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, as amended, and Section 21E of the Securities Exchange Act of 1934, as amended, which statements involve substantial risks and uncertainties. Forward-looking statements generally relate to future events or our future financial or operating performance. In some cases, you can identify forward-looking statements because they contain words such as “may,” "aim," “will,” “should,” “expect,” “plan,” "try," “anticipate,” “could,” “would,” “intend,” “target,” “project,” “contemplate,” “believe,” “estimate,” “predict,” “potential” or “continue” or the negative of these words or other similar terms or expressions that concern our expectations, strategies, plans, or intentions. Forward-looking statements in this shareholder letter include, but are not limited to: our expectations regarding our financial position and operating performance; our expectations regarding our products, systems and platform innovation, including the development of our Autonomous Delivery Platform and global technology platform and our usage of AI; our plans and expectations regarding our overall business strategy and investment approach; our operational philosophy; our strategies and expectations regarding improvements to our products and systems; our expectations regarding our growth prospects and our local commerce opportunity; trends in our business; and demand for our platform and for local commerce platforms in general. Our expectations and beliefs regarding these matters may not materialize, and actual results in future periods are subject to risks and uncertainties that could cause actual results to differ materially from those projected, including risks and uncertainties related to: economic, financial, social or political conditions that could adversely affect us; competition; managing our growth and corporate culture; the macroeconomic environment and geopolitical uncertainty; financial performance; investments in new geographies, products, or offerings, as well as our technology infrastructure; our ability to successfully integrate and realize the benefits of acquisitions, including Deliveroo, strategic partnerships, joint ventures, and investments; our ability to attract merchants, consumers, and Dashers to our platform; legal proceedings and regulatory matters and developments; any future changes to our business or our financial or operating model; and our brand and reputation. The forward-looking statements contained in this shareholder letter are also subject to other risks and uncertainties that could cause actual results to differ from the results predicted, including those more fully described in our filings with the SEC, including our Annual Report on Form 10-K for the year ended December 31, 2025. All forward-looking statements in this shareholder letter are based on information available to DoorDash and assumptions and beliefs as of the date hereof, and we disclaim any obligation to update any forward-looking statements, except as required by law.