AMBARELLA INC (AMBA)

ITEM 2. Management’s Discussion and Analysis of Financial Condition and Results of Operations

The following discussion and analysis of our financial condition and results of operations should be read in conjunction with the unaudited condensed consolidated financial statements and related notes included elsewhere in this Quarterly Report on Form 10-Q, and the consolidated financial statements and notes thereto for the fiscal year ended January 31, 2026 and management’s discussion and analysis of our financial condition and results of operations included in our Annual Report on Form 10-K for the 2026 fiscal year filed with the Securities and Exchange Commission, or SEC, on March 23, 2026.

This Quarterly Report on Form 10-Q, including this “Management’s Discussion and Analysis of Financial Condition and Results of Operations”, includes a number of forward-looking statements that involve many risks and uncertainties. Forward-looking statements are identified by the use of the words “would,” “could,” “will,” “may,” “expect,” “believe,” “should,” “anticipate,” “outlook,” “if,” “future,” “intend,” “plan,” “estimate,” “predict,” “potential,” “target,” “seek,” “project,” “forecast,” “continue” or “foreseeable” and similar words and phrases, including the negatives of these terms, or other variations of these terms, that denote future events. Such statements include, but are not limited to, statements concerning our market opportunity and our ability to compete in such markets, our product strategy, our ability to develop and introduce new solutions, our future financial and operating performance, our sales and marketing strategy, our investment strategy, research and development, our customer and supplier relationships and inventory levels, industry trends, our cash needs and capital requirements, our repurchase programs, our expectations about taxes, operating expenses, and cost recognition, the availability of third-party components and economic and political conditions. These statements reflect our current views with respect to future events and our potential financial performance, and are subject to risks and uncertainties that could cause our actual results and financial position to differ materially and adversely from what is projected or implied in any forward-looking statements included in this Quarterly Report on Form 10-Q. These factors include, but are not limited to: risks associated with revenue being generated from new customers or design wins, neither of which is assured; our ability to retain and expand customer relationships and to achieve design wins; risks associated with the overall economy, including higher inflation and trade tensions between the U.S. and China; the commercial success of our customers’ products; our growth strategy; fluctuations in our operating results; our ability to anticipate future market demands and future needs and preferences of our customers; our ability to introduce new and enhanced solutions; the expansion of our current markets and our ability to successfully enter new markets; anticipated trends and challenges, including competition, in the markets in which we operate or seek to operate; our expectations regarding artificial intelligence and computer vision; our ability to effectively generate and manage growth; our ability to retain key employees; the potential for intellectual property disputes or other litigation; the risks described under Item 1A of Part II — “Risk Factors,” and Item 2 of Part I — “Management’s Discussion and Analysis of Financial Condition and Results of Operations”; the risks described elsewhere in this Quarterly Report on Form 10-Q and those discussed in other documents we file with the SEC. We make these forward-looking statements based upon information available on the date of this Quarterly Report on Form 10-Q, and we have no obligation (and expressly disclaim any such obligation) to update or alter any forward-looking statements, whether as a result of new information or otherwise except as otherwise required by securities regulations.

Overview

We are a leading developer of low power system on a chip (SoC) semiconductors and software that enable advanced edge and physical AI applications. Our solutions combine state of the art video processing, high resolution image capture, and our proprietary CVflow® AI acceleration architecture to deliver high performance at extremely low power. Historically, our technologies supported human viewing applications such as enterprise, public infrastructure, and home security cameras, as well as sports cameras, wearables, aerial drones, and aftermarket automotive recorders. Building on this foundation, our recent product generations incorporate advanced AI inference capabilities that allow edge devices to interpret complex scenes, perform multi modal sensor fusion, and support autonomous decision making.

Our latest SoC families integrate third generation CVflow technology with advanced video processing, image signal processing, audio processing, and system control functions on a single chip. CVflow is optimized for a broad range of AI inference workloads, including object detection, classification, tracking, segmentation, stereo depth processing, radar perception, and transformer based models. This architecture supports multi modal sensor inputs, including camera, lidar, 4D radar, thermal, and near infrared, enabling environmental perception for edge devices. These capabilities allow our customers to deploy differentiated AI models and solutions across applications, such as next generation automotive camera systems, video security, robotics, and consumer devices, while achieving high image quality, low latency, and low power consumption.

20

Our development roadmap is focused on human viewing, AI inference, and radar based perception technologies that support the increasing automation and intelligence requirements of the Internet of Things (IoT), automotive, industrial, and robotics markets. As a result, we believe that our future revenue growth, if any, will significantly depend upon our ability to expand within camera markets with our AI technology, particularly in the Internet of Things, or IoT, markets, as well as emerging markets such as AI-enabled security cameras, AI-based driving applications, including driver monitoring systems, advanced blind spot detection, object detection, and deep learning algorithms for HD mapping solutions, automotive advanced driver assistance systems, or ADAS, applications, and industrial and robotics markets. We expect our research and development expenditures to increase in comparison to prior periods as we devote additional resources to the development of innovative AI inference SoCs with increased functionality, and as we target new markets.

We sell our SoC solutions to leading original design manufacturers, or ODMs, and original equipment manufacturers, or OEMs, globally, and in the automotive market, we also sell to Tier-1 suppliers. We refer to ODMs and Tier-1 automotive suppliers as our customers and OEMs as our end customers, except as otherwise indicated or as the context otherwise requires.

Our sales cycles typically require a significant investment of time and a substantial expenditure of resources before we can realize revenue from the sale of our solutions, if any. Our typical sales cycle consists of a multi-month sales and development process involving our customers’ system designers and management and our sales personnel and software engineers. If successful, this process culminates in a customer’s decision to use our solutions in its system, which we refer to as a design win. Our sales efforts are typically directed to the OEM of the product that will incorporate our video and image processing solution, but the eventual design and incorporation of our SoC into the product may be handled by an ODM or Tier-1 supplier on behalf of the OEM.

Volume production may begin within 12 to 18 months after a design win, but could be longer in certain markets, depending on the complexity of our customer’s product and other factors upon which we may have little or no influence. In general, design cycles will be longer in the OEM automotive and industrial and robotics markets than in the IoT markets. Once our solutions have been incorporated into a customer’s design, they are likely to be used for the life cycle of the customer’s product. Conversely, a design loss to a competitor will likely preclude any opportunity for future revenue from such customer’s product. Even if we obtain a design win and our SoC remains a component through the life cycle of a customer’s product, the volume and timing of actual sales of our SoCs to the customer depend upon the production, release and market acceptance of that product, none of which are within our control. An IoT product typically has a life cycle of 12 to 24 months. We anticipate that product life cycles will typically be longer than 24 months in the OEM automotive and industrial and robotics markets, as new product introductions occur less frequently in these markets.

Financial Highlights

•
We recorded revenue of $100.4 million for the three months ended April 30, 2026, an increase of 16.9% as compared to the same period in the prior fiscal year. The increase in revenue was primarily attributable to higher product unit shipments and an increased percentage of sales from higher average selling price AI inference processors as a result of high demand for our edge AI solutions, partially offset by lower nonrecurring engineering (NRE) project service revenue.

•
We recorded operating losses of $19.4 million for the three months ended April 30, 2026, as compared to operating losses of $25.9 million for the three months ended April 30, 2025. The reduction in operating losses was primarily attributable to higher revenue and, consequently, higher gross profit, partially offset by higher operating expenses. The increase in operating expenses primarily resulted from higher payroll-related expenses as a result of an increase in headcount, higher engineering-related costs associated with the progress and number of chips in development, and higher marketing expense, partially offset by lower stock-based compensation expense.

•
We had cash outflows from operating activities of $25.6 million for the three months ended April 30, 2026, as compared to cash inflows of $14.8 million for the three months ended April 30, 2025. The decrease in cash flows from operating activities was mainly driven by higher cash outflows from changes in working capital, primarily due to increased inventory purchases, partially offset by improved operating results.

•
During the three months ended April 30, 2026, we repurchased a total of 47,798 shares for approximately $2.4 million in cash. The repurchased shares were recorded as authorized but unissued shares. As of April 30, 2026, there was approximately $45.6 million available for repurchases under the existing repurchase program through June 30, 2026.

Factors Affecting Our Performance

21

Ability to Capitalize on AI and Computer Vision Trends. We expect that AI and computer vision functionality will become an increasingly important requirement in many of our current and future markets, including IoT, automotive, industrial and robotics markets. As a result, we believe that our ability to develop advanced AI and computer vision technologies, enable and support customer product development in emerging applications, such as ADAS, advanced blind spot detection, object detection, classification and tracking, people recognition, retail analytics, and machine learning, and gain customer acceptance of our technology platform and solutions will be critical to our future success. Moreover, achieving design wins, particularly for computer vision-centric applications in the IoT, automotive, industrial and robotics markets, is vital to our ability to generate revenue growt

[Excerpt truncated for page length; source filing is linked above.]

ITEM 7. MANAGEMENT’S DISCUSSION AND ANALYSIS OF FINANCIAL CONDITION AND RESULTS OF OPERATIONS

Overview

We are a leading developer of low power system on a chip (SoC) semiconductors and software that enable advanced edge and physical AI applications. Our solutions combine state of the art video processing, high resolution image capture, and our proprietary CVflow® AI acceleration architecture to deliver high performance at extremely low power. Historically, our technologies supported human viewing applications such as enterprise, public infrastructure, and home security cameras, as well as sports cameras, wearables, aerial drones, and aftermarket automotive recorders. Building on this foundation, our recent product generations incorporate advanced AI inference capabilities that allow edge devices to interpret complex scenes, perform multi modal sensor fusion, and support autonomous decision making.

Our latest SoC families integrate third generation CVflow technology with advanced video processing, image signal processing, audio processing, and system control functions on a single chip. CVflow is optimized for a broad range of AI inference workloads, including object detection, classification, tracking, segmentation, stereo depth processing, radar perception, and transformer based models. This architecture supports multi modal sensor inputs, including camera, lidar, 4D radar, thermal, and near infrared, enabling environmental perception for edge devices. These capabilities allow our customers to deploy differentiated AI models and solutions across applications, such as next generation automotive camera systems, video security, robotics, and consumer devices, while achieving high image quality, low latency, and low power consumption.

Our development roadmap is focused on human viewing, AI inference, and radar based perception technologies that support the increasing automation and intelligence requirements of the Internet of Things (IoT), automotive, industrial, and robotics markets. As a result, we believe that our future revenue growth, if any, will significantly depend upon our ability to expand within camera markets with our AI technology, particularly in the Internet of Things, or IoT, markets, as well as emerging markets such as AI-enabled security cameras, AI-based driving applications, including driver monitoring systems, advanced blind spot detection, object detection, and deep learning algorithms for HD mapping solutions, automotive advanced driver assistance systems, or ADAS, applications, and industrial and robotics markets. We expect our research and development expenditures to increase in comparison to prior periods as we devote additional resources to the development of innovative video and image processing solutions with increased functionality, such as AI capabilities, and as we target new markets.

We sell our SoC solutions to leading original design manufacturers, or ODMs, and original equipment manufacturers, or OEMs, globally, and in the automotive market, we also sell to Tier-1 suppliers. We refer to ODMs and Tier-1 automotive suppliers as our customers and OEMs as our end customers, except as otherwise indicated or as the context otherwise requires.

Our sales cycles typically require a significant investment of time and a substantial expenditure of resources before we can realize revenue from the sale of our solutions, if any. Our typical sales cycle consists of a multi-month sales and development process involving our customers’ system designers and management and our sales personnel and software engineers. If successful, this process culminates in a customer’s decision to use our solutions in its system, which we refer to as a design win. Our sales efforts are typically directed to the OEM of the product that will incorporate our video and image processing solution, but the eventual design and incorporation of our SoC into the product may be handled by an ODM or Tier-1 supplier on behalf of the OEM.

Volume production may begin within 12 to 18 months after a design win, but could be longer in certain markets, depending on the complexity of our customer’s product and other factors upon which we may have little or no influence. In general, design cycles will be longer in the OEM automotive and industrial and robotics markets than in the IoT markets. Once our solutions have been incorporated into a customer’s design, they are likely to be used for the life cycle of the customer’s product. Conversely, a design loss to a competitor will likely preclude any opportunity for future revenue from such customer’s product. Even if we obtain a design win and our SoC remains a component through the life cycle of a customer’s product, the volume and timing of actual sales of our SoCs to the customer depend upon the production, release and market acceptance of that product, none of which are within our control. An IoT product typically has a life cycle of 12 to 24 months. We anticipate that product life cycles will typically be longer than 24 months in the OEM automotive and industrial and robotics markets, as new product introductions occur less frequently in these markets.

50

Fiscal Year 2026 Financial Highlights

•
We recorded revenue of $390.7 million in fiscal year 2026, an increase of 37.2% as compared to fiscal year 2025. The increase in revenue was primarily attributable to higher product unit shipments and an increased percentage of sales from higher average selling price AI inference processors as a result of high demand for our edge AI solutions, partially offset by lower nonrecurring engineering (NRE) project service revenue.

•
We recorded a loss from operations of $82.5 million in fiscal year 2026, as compared to a loss from operations of $126.6 million in fiscal year 2025. The reduction in operating losses was primarily attributable to higher revenue and consequently, higher gross profit, partially offset by an increase in operating expenses. The increased operating expenses primarily related to higher personnel costs, including employee benefits, higher engineering-related costs associated with the progress and number of chips in development, higher facility-related costs from relocation of our headquarters in fiscal year 2026, as well as higher marketing and travel expenses.

•
We generated $73.5 million of cash from operating activities in fiscal year 2026, as compared to $33.8 million in fiscal year 2025. The increase in cash flows from operating activities was mainly driven by improved operating results and higher cash inflows from changes in working capital.

Factors Affecting Our Performance

Ability to Capitalize on AI and Computer Vision Trends. We expect that AI and computer vision functionality will become an increasingly important requirement in many of our current and future markets, including IoT, automotive, industrial and robotics markets. As a result, we believe that our ability to develop advanced AI and computer vision technologies, enable and support customer product development in emerging applications, such as ADAS, advanced blind spot detection, object detection, classification and tracking, people recognition, retail analytics, and machine learning, and gain customer acceptance of our technology platform and solutions will be critical to our future success. Moreover, achieving design wins, particularly for computer vision-centric applications in the IoT, automotive, industrial and robotics markets, is vital to our ability to generate revenue growth. As such, we closely monitor our design wins with our customers. However, a design win may not successfully materialize into revenue, and even if it does result in revenue, the amount generated by each design win can vary significantly.

Ability to Develop and Introduce New or Enhanced Solutions. We operate in a dynamic environment characterized by rapidly changing technologies and technological obsolescence. To compete successfully, we must design, develop, market and sell enhanced solutions with increased levels of performance and functionality that meet the expectations of our customers, including advanced process technologies. As such, we continuously invest in our research and development projects, especially AI and computer vision technologies. However, failure to anticipate or timely develop new or enhanced solutions in response to technology shifts and trends could result in decreased revenue and our competitors achieving design wins we sought. In addition, our ability to successfully develop new and enhanced solutions depends on our continuing ability to hire, train, motivate and retain highly skilled engineers, which is not ensured. Moreover, any reliability or quality problems with our solutions could harm our reputation, increase additional development and replacement costs, and prevent us from retaining existing customers and attracting new customers.

Pricing, Product Cost and Margin. Our pricing and margins depend on a variety of factors, including the volumes and features of the solutions we provide to our customers. Additionally, we make significant investments in new solutions for both cost improvements and new features that we expect to drive revenue and maintain margins. In general, solutions incorporated into more complex configurations, such as those used in high-performance camera applications or, in the future, advanced driver assistance systems, have higher prices and higher gross margins as compared to solutions sold into lower-performing, more competitive camera applications. Our average selling price can vary by market and application due to market-specific supply and demand, the maturation of products launched in previous years, the launch of new products by us or our competitors and by product mix.

We continually monitor the cost of our solutions. As we rely on third-party manufacturers for the manufacture of our products, we maintain a close relationship with these suppliers to continually monitor production yields, component costs and design efficiencies.

51

Continued Concentration of Revenue by End Markets. Historically, our revenue has been significantly concentrated in a small number of end markets and we developed technologies to provide solutions for new markets as they emerged. Since fiscal year 2018, the IoT markets and automotive markets have been our largest end markets and sales into these markets collectively generated the majority of our revenue. We believe, however, that continued expansion into new markets is required to facilitate revenue growth and customer diversification. We have recently introduced solutions to address emerging applications and markets, such as the incorporation of AI and computer vision functionalities for AI-enabled security cameras, AI-based driving applications and industrial and robotics markets. While we will continue to seek to expand our end market exposure, we anticipate that sales to a limited number of markets will continue to account for a significant percentage of our total revenue for the foreseeable future. Our concentration in a limited number of markets may cause our financial performance to fluctuate significantly from period to period based on the success or failure of products that our SoCs are designed into as well as the overall growth or decline in the video capture markets in which we compete. In addition, we derive a significant portion of our revenue from a limited number of ODMs who build products on behalf of a limited number of OEMs and from a limited number of OEMs to whom we ship directly. We believe that our operating results for the foreseeable future will continue to depend on sales to a relatively small number of customers.

Sales Volume. A typical design win that successfully launches into the marketplace can generate a wide range of sales volumes for our solutions, depending on the end market demand for our customers’ products. Our ability to accurately forecast demand

[Excerpt truncated for page length; the complete text is on the linked full-MD&A page.]