Grid Dynamics Holdings, Inc. is an Enterprise Artificial Intelligence transformation partner for the Fortune 1000. The company combines deep AI expertise with enterprise scale delivery to help clients identify where to invest in AI build systems that operate at scale and capture measurable business value from AI deployments. Its technical foundation spans distributed systems real time data machine learning algorithms and natural language processing which have converged into…
Grid Dynamics Holdings, Inc. is an Enterprise Artificial Intelligence transformation partner for the Fortune 1000. The company combines deep AI expertise with enterprise scale delivery to help clients identify where to invest in AI build systems that operate at scale and capture measurable business value from AI deployments. Its technical foundation spans distributed systems real time data machine learning algorithms and natural language processing which have converged into what the firm calls Enterprise AI. Founded in 2006 and headquartered in Silicon Valley the firm leverages nearly two decades of technology leadership to serve large enterprises seeking custom engineered AI solutions.
Grid Dynamics generates revenue by providing a suite of professional services that span artificial intelligence cloud platform and product engineering digital engagement Internet of Things and edge computing and data and AI platforms. The firm’s AI services guide clients through readiness assessments use case identification robust infrastructure deployment and enterprise wide scaling. Cloud platform and product engineering modernizes legacy systems and builds cloud native platforms that support continuous delivery and AI integration. Digital engagement delivers composable commerce and customer experience solutions that embed AI throughout the buying journey. Internet of Things and edge computing offerings enable manufacturers and logistics providers to implement physical AI for operational safety and intelligence. Data and AI platform services create the infrastructure needed for scalable analytics machine learning operations and real time streaming. Customers are primarily Fortune 1000 enterprises that seek measurable outcomes such as increased revenue reduced costs or improved satisfaction.
The company operates through the following industry focused verticals: Retail, Technology, Media and Telecom, Finance, CPG/Manufacturing, Healthcare and Pharma, and Other.
• Retail: This segment provides AI powered discovery through visual search, conversational AI, and intelligent merchandising; composable commerce with headless architectures; omnichannel fulfillment including real time inventory, intelligent routing, and last mile optimization; and physical AI for in store robotics, computer vision, and digital twin simulation.
• Technology Media and Telecom: This segment serves technology native clients that demand cutting edge AI capabilities at scale including technology companies, media enterprises, and telecommunications providers with cloud platform engineering, AI native application development, data platforms, and scalability solutions.
• Finance: This segment serves banks, insurance companies, wealth management firms, and fintech innovators with solutions spanning AI powered bitemporal analytics, agentic regulatory compliance automation, customer experience platforms, financial advisor AI copilots, and core banking modernization.
• CPG Manufacturing: This segment serves CPG companies and manufacturers with AI powered operations solutions including smart manufacturing, supply chain optimization platforms, IoT implementations, anomaly detection, predictive maintenance, visual process monitoring, digital twins, and collaborative robotics applications.
• Healthcare and Pharma: This segment serves leading companies with cloud app modernization, data science, and digital commerce solutions to accelerate drug development, product development, and time to market, optimize marketing and sales operations, and personalize customer experiences with data insights and automation.
• Other: This segment serves clients in other sectors where AI and engineering capabilities create value requiring domain specific solutions.
Grid Dynamics positions itself as a leader in the enterprise AI services market by emphasizing its AI native heritage deep technical expertise and a delivery model that prioritizes senior engineering talent over volume staffing. The firm differentiates itself from generalist consultancies and traditional IT services providers through its focus on production grade AI engineering proprietary solution accelerators and strategic alliances with hyperscale cloud providers and AI technology leaders such as Google Cloud AWS Microsoft Azure NVIDIA and Temporal Technologies. Competitors include large global IT services firms like Accenture EPAM and Capgemini as well as specialized AI consultancies and offshore providers such as Cognizant Infosys TCS and Wipro. The company’s competitive strengths lie in its combination of deep AI expertise proven frameworks a culture of engineering excellence and a global delivery model that enables round the clock development.
Grid Dynamics serves Fortune 1000 companies across multiple industries with a concentration of revenue among a limited number of large accounts. In the year ended December 31, 2025 a single customer accounted for 15.4% of total revenue and the ten largest customers together contributed 57.7% of revenue. The firm’s customer base includes major retailers, technology and telecom enterprises, banks and financial institutions, consumer packaged goods manufacturers, and healthcare and pharmaceutical companies. While the filing does not disclose specific customer names it indicates that the company’s vertical focus covers retail, technology media and telecom, finance, CPG manufacturing, and healthcare and pharma sectors.
Sector:TechnologySector rationaleGrid Dynamics is primarily an IT services firm that designs and builds custom AI solutions, cloud platforms, and data infrastructure for Fortune 1000 clients, which falls under the 'IT Services' industry within the Technology sector. A secondary sector of Industrials is justified because the company provides specific 'Internet of Things and edge computing' and 'physical AI' services for manufacturers and logistics providers, including smart manufacturing and collaborative robotics applications.Industry:IT ServicesTechnologyPrimaryGrid Dynamics provides professional services including custom engineered AI solutions, cloud platform and product engineering, and digital transformation engagements for Fortune 1000 clients. Its revenue model is based on providing technology labor and expertise as a service, competing directly with other IT services firms like Accenture, EPAM, and Capgemini.Classified using BQ-MICSCIK: 0001743725
Investment Thesis
▲ Bull case
Grid Dynamics is positioned to capture significant value from its AI-native delivery model, which is fundamentally altering its cost structure and revenue potential. The company reported that AI revenue reached 29.3% of total revenue in Q1 FY26, growing nearly 60% year-over-year, signaling that its AI practice is no longer experimental but core to the business. This concentration enables structural margin expansion, as AI-native delivery reduces reliance on labor-intensive time-and-materials models and shifts toward outcome-based, fixed-price engagements. Internal metrics show AI workflows at a global bank analyzing 150 green production applications, uncovering latent defects and expanding validated behavior coverage beyond 70%, which directly reduced false confidence in system integrity and mitigated production security and regulatory risk. This level of technical precision and risk mitigation is not merely an efficiency gain—it represents a qualitative leap in service delivery that justifies premium pricing and strengthens client retention. The company’s ability to deliver measurable ROI at commercial scale, as evidenced by 50% reduction in proposal preparation time and 18% increase in monthly spend at a major food distributor, demonstrates that AI is driving revenue-generating outcomes for clients, not just cost savings. This creates a virtuous cycle where successful deployments lead to expanded scope, longer contract durations, and higher wallet share—trends already visible in retention data and expansion of existing hyperscaler co-sell accounts. The market is underestimating how deeply this model is embedded in client operations, with the top 5 accounts now entirely outside retail and dominated by technology and financial services firms undergoing vendor consolidation where Grid Dynamics has emerged as a clear beneficiary. This vertical shift toward high-AI-adoption sectors, combined with the company’s product-centric engineering approach, positions it to win larger, more strategic deals that were previously out of reach due to legacy delivery model constraints.
The GAIN platform strategy, particularly its deployment on hyperscaler marketplaces, is creating a scalable, high-margin go-to-market engine that is still in early stages of monetization. Partner inference revenues already constitute 19.1% of total revenue in Q1 FY26, with leadership targeting 25-30% long-term—indicating significant runway for growth. The recent announcement of the AI-native modernization service on Microsoft Azure, powered by the GAIN Platform for SDLC, is a direct embodiment of this strategy and targets a high-value market segment: Fortune 1000 enterprises with mission-critical, high-transaction-volume legacy environments. Azure’s revenue surpassed $75 billion in 2025, growing 34% year-over-year, reflecting accelerating enterprise migration demand that Grid Dynamics is uniquely positioned to capture as a Microsoft Azure specialized partner with five advanced specializations. Through the Azure Accelerate program, clients receive free deployment assistance, Azure credits, and funded migration assessments—removing financial and technical barriers that typically stall large-scale modernization. The GAIN Platform for SDLC internally benchmarks to accelerate project delivery by over 30% and enables outcome-based pricing, directly tying compensation to performance. This model compresses migration timelines and transforms legacy modernization from a cost center into a strategic, innovation-enabling initiative. Unlike traditional services engagements, these platform-led deals generate higher gross margins due to reduced labor intensity, IP leverage, and stickiness from embedded Forward Deployed Engineers. The platform’s GenAI-powered data migration automation converts legacy SQL, pipeline orchestration, and data schemas from Teradata, Informatica, and Oracle directly to Azure-native equivalents, eliminating multi-year timelines and cost overruns. This addresses a critical pain point—technical debt and legacy licensing costs—that enterprises have long deemed too expensive or risky to tackle. By solving this, Grid Dynamics is not just migrating systems but enabling fundamental transformation, unlocking massive innovation potential. The market is overlooking how this specific offering converts a historically low-margin, high-risk legacy modernization business into a recurring, high-margin AI-driven revenue stream with strong expansion potential across AWS and Google Cloud marketplaces.
Grid Dynamics’ internal AI automation initiatives are creating a hidden lever for operational leverage that will drive margin expansion independent of revenue growth. The company reported a 2x productivity improvement in recruitment processing, a 50% increase in RFP responses without growing headcount, knowledge management response times improved from hours to minutes, and HR initiatives targeting over 20% operational improvement. These gains are not anecdotal—they reflect systemic adoption of AI tools across core corporate functions, reducing friction and increasing throughput. The fourth pillar of its AI transformation—internal AI automation—is not merely a cost-saving exercise; it is building organizational muscle memory for AI integration that will enhance delivery capacity and client responsiveness. As internal teams become more efficient at onboarding talent, responding to opportunities, and managing knowledge, the company can handle more complex engagements without proportional headcount growth. This directly supports the revenue per person metric that management identified as a key utilization gauge. The implication is that Grid Dynamics can scale its addressable market and take on larger, more complex AI-native projects—such as full mainframe modernization at a leading home improvement retailer or enabling intelligent autonomous capabilities in mining equipment—without linearly increasing costs. This operational leverage is especially critical as the company shifts toward fixed-price, outcome-based contracts where margin is determined by efficiency, not hours billed. The market is failing to recognize that these internal efficiencies are not one-time gains but are compounding as AI tools mature and become embedded in workflows. With headcount remaining flat year-over-year at 4,964 despite strong revenue growth in high-margin verticals like TMT (up 30.3% YoY) and Finance (23.5% of revenue), the underlying productivity gains are already materializing. As these internal AI capabilities mature, they will enable higher utilization, better project execution, and improved gross and EBITDA margins—creating a self-reinforcing cycle of efficiency and profitability that is not yet reflected in current valuations.
Grid Dynamics is positioned to capture significant value from its AI-native delivery model, which is fundamentally altering its cost structure and revenue potential. The company reported that AI revenue reached 29.3% of total revenue in Q1 FY26, growing nearly 60% year-over-year, signaling that its AI practice is no longer experimental but core to the business. This concentration enables structural margin expansion, as AI-native delivery reduces reliance on labor-intensive time-and-materials models and shifts toward outcome-based, fixed-price engagements. Internal metrics show AI workflows at a global bank analyzing 150 green production applications, uncovering latent defects and expanding validated behavior coverage beyond 70%, which directly reduced false confidence in system integrity and mitigated production security and regulatory risk. This level of technical precision and risk mitigation is not merely an efficiency gain—it represents a qualitative leap in service delivery that justifies premium pricing and strengthens client retention. The company’s ability to deliver measurable ROI at commercial scale, as evidenced by 50% reduction in proposal preparation time and 18% increase in monthly spend at a major food distributor, demonstrates that AI is driving revenue-generating outcomes for clients, not just cost savings. This creates a virtuous cycle where successful deployments lead to expanded scope, longer contract durations, and higher wallet share—trends already visible in retention data and expansion of existing hyperscaler co-sell accounts. The market is underestimating how deeply this model is embedded in client operations, with the top 5 accounts now entirely outside retail and dominated by technology and financial services firms undergoing vendor consolidation where Grid Dynamics has emerged as a clear beneficiary. This vertical shift toward high-AI-adoption sectors, combined with the company’s product-centric engineering approach, positions it to win larger, more strategic deals that were previously out of reach due to legacy delivery model constraints.
The GAIN platform strategy, particularly its deployment on hyperscaler marketplaces, is creating a scalable, high-margin go-to-market engine that is still in early stages of monetization. Partner inference revenues already constitute 19.1% of total revenue in Q1 FY26, with leadership targeting 25-30% long-term—indicating significant runway for growth. The recent announcement of the AI-native modernization service on Microsoft Azure, powered by the GAIN Platform for SDLC, is a direct embodiment of this strategy and targets a high-value market segment: Fortune 1000 enterprises with mission-critical, high-transaction-volume legacy environments. Azure’s revenue surpassed $75 billion in 2025, growing 34% year-over-year, reflecting accelerating enterprise migration demand that Grid Dynamics is uniquely positioned to capture as a Microsoft Azure specialized partner with five advanced specializations. Through the Azure Accelerate program, clients receive free deployment assistance, Azure credits, and funded migration assessments—removing financial and technical barriers that typically stall large-scale modernization. The GAIN Platform for SDLC internally benchmarks to accelerate project delivery by over 30% and enables outcome-based pricing, directly tying compensation to performance. This model compresses migration timelines and transforms legacy modernization from a cost center into a strategic, innovation-enabling initiative. Unlike traditional services engagements, these platform-led deals generate higher gross margins due to reduced labor intensity, IP leverage, and stickiness from embedded Forward Deployed Engineers. The platform’s GenAI-powered data migration automation converts legacy SQL, pipeline orchestration, and data schemas from Teradata, Informatica, and Oracle directly to Azure-native equivalents, eliminating multi-year timelines and cost overruns. This addresses a critical pain point—technical debt and legacy licensing costs—that enterprises have long deemed too expensive or risky to tackle. By solving this, Grid Dynamics is not just migrating systems but enabling fundamental transformation, unlocking massive innovation potential. The market is overlooking how this specific offering converts a historically low-margin, high-risk legacy modernization business into a recurring, high-margin AI-driven revenue stream with strong expansion potential across AWS and Google Cloud marketplaces.
Grid Dynamics’ internal AI automation initiatives are creating a hidden lever for operational leverage that will drive margin expansion independent of revenue growth. The company reported a 2x productivity improvement in recruitment processing, a 50% increase in RFP responses without growing headcount, knowledge management response times improved from hours to minutes, and HR initiatives targeting over 20% operational improvement. These gains are not anecdotal—they reflect systemic adoption of AI tools across core corporate functions, reducing friction and increasing throughput. The fourth pillar of its AI transformation—internal AI automation—is not merely a cost-saving exercise; it is building organizational muscle memory for AI integration that will enhance delivery capacity and client responsiveness. As internal teams become more efficient at onboarding talent, responding to opportunities, and managing knowledge, the company can handle more complex engagements without proportional headcount growth. This directly supports the revenue per person metric that management identified as a key utilization gauge. The implication is that Grid Dynamics can scale its addressable market and take on larger, more complex AI-native projects—such as full mainframe modernization at a leading home improvement retailer or enabling intelligent autonomous capabilities in mining equipment—without linearly increasing costs. This operational leverage is especially critical as the company shifts toward fixed-price, outcome-based contracts where margin is determined by efficiency, not hours billed. The market is failing to recognize that these internal efficiencies are not one-time gains but are compounding as AI tools mature and become embedded in workflows. With headcount remaining flat year-over-year at 4,964 despite strong revenue growth in high-margin verticals like TMT (up 30.3% YoY) and Finance (23.5% of revenue), the underlying productivity gains are already materializing. As these internal AI capabilities mature, they will enable higher utilization, better project execution, and improved gross and EBITDA margins—creating a self-reinforcing cycle of efficiency and profitability that is not yet reflected in current valuations.
Grid Dynamics’ heavy reliance on a concentrated customer base poses a significant and underappreciated risk to revenue stability, despite management’s emphasis on diversification. While the company highlights that its top 5 accounts are now entirely outside retail and include two global technology firms, a fintech leader, a U.S.-based global bank, and a leading financial institution, this concentration creates vulnerability. In Q1 FY26, revenues from the top 5 and top 10 customers reached 40.8% and 59.7%, respectively—up from 35.6% and 56.6% a year ago—indicating increasing, not decreasing, reliance on a small number of clients. The loss of any single top-tier account due to strategic shifts, budget cuts, or insourcing could disproportionately impact revenue. Furthermore, the company’s narrative around vendor consolidation benefiting Grid Dynamics assumes these clients will continue to expand their relationship, but there is no guarantee that consolidation will favor Grid Dynamics over other specialized AI vendors or that these enterprises won’t eventually bring critical AI workloads in-house as they build internal capabilities. The financial services and technology verticals, while currently strong, are also subject to intense competition from larger system integrators and niche AI specialists who may offer deeper domain expertise or broader platform ecosystems. Management’s confidence in being a “preferred vendor” is based on current engagement depth, but this status is not contractually locked and could erode if clients perceive better value elsewhere or if macroeconomic pressures trigger renewed cost-cutting initiatives. The market may be ignoring how fragile this preferred status is in an environment where enterprises are rapidly building internal AI centers of excellence and reevaluating external partnerships based on strict ROI thresholds.
The transition to AI-native, outcome-based, and fixed-price contracts introduces significant execution and financial risks that management has not adequately addressed, particularly regarding revenue recognition, margin volatility, and project overruns. While Eugene Steinberg highlighted the use of AI agents and the Rosetta framework to uncover requirement uncertainties during presale, and Anil Doradla acknowledged past learnings from fixed-price engagements pre-AI, the company admits it is still “experimenting” with monetization timing and revenue recognition for these new models. The shift from time-and-materials to fixed-bid work means revenue is tied to milestone achievement or outcome validation, not hours delivered—creating potential for revenue deferral, recognition delays, or even reversals if outcomes fall short of agreed criteria. Anil noted that the company is “taking baby steps” and that the optics of guidance look different from underlying business realities due to nonlinear timing of monetization. This uncertainty is compounded by the fact that Grid Dynamics is moving into higher-complexity domains like physical AI and mainframe modernization, where technical unknowns are more prevalent. For example, while the company successfully modernized a legacy mainframe platform for a home improvement retailer using AI agents, such projects involve high stakes—failure could result in significant reputational damage, financial penalties, or legal exposure if system integrity or regulatory compliance is compromised. The Rosetta framework and AI coding assistance may reduce uncertainty, but they do not eliminate it, especially in legacy systems with poor documentation or complex interdependencies. The market may be underestimating the likelihood of scope creep, technical surprises, or client disagreements over outcome metrics—particularly in early-stage AI deployments where success criteria are still evolving. If Grid Dynamics fails to consistently deliver on outcome-based promises, it could face margin compression, disputed invoices, damaged client trust, and a slower-than-expected ramp in non-T&M revenue contribution—undermining the very thesis that justifies its premium valuation.
Grid Dynamics’ operating model remains structurally challenged by persistent FX headwinds and rising cost pressures in its global delivery footprint, which are eroding margins despite growth in high-potential verticals. Anil Doradla explicitly cited FX fluctuations as a headwind of approximately $1.2 million on a year-over-year basis in Q1 FY26, impacting EBITDA, and attributed sequential and year-over-year EBITDA declines to a combination of FX headwinds and higher operating costs across delivery locations. The company’s non-U.S. headcount stands at 92.9% (4,611 employees), with significant exposure to Europe, Latin America, and India—regions where currency volatility and wage inflation are persistent. While the company utilizes natural hedges and an active hedging program, these are reactive measures that do not eliminate structural exposure. Furthermore, the shift toward AI-native delivery and productized engineering requires upskilling and retraining of talent, which increases labor costs in the short term. Although management cites internal AI automation gains (e.g., 2x recruitment productivity, 50% more RFP responses), these efficiencies are still nascent and may not yet offset the cost burden of retraining, geographic reorganization, and transaction-related expenses. The GAAP gross margin declined year-over-year from 36.8% to 34.8%, and non-GAAP gross profit margin fell from 37.4% to 35.3%, reflecting real pressure on core profitability. Management’s guidance for improved margins hinges on the success of new platform-based engagements, but if AI monetization lags or fixed-price projects underperform due to execution risks, the company could be caught in a squeeze: higher delivery costs from talent transformation without commensurate revenue uplift. The market may be overlooking how these structural cost pressures—exacerbated by geographic concentration in lower-wage but inflation-prone regions and currency volatility—could persist longer than anticipated, limiting margin expansion even as revenue grows in AI. Without clear evidence that internal AI automation is driving sustained, scalable cost reductions across delivery operations—not just support functions—the margin expansion story remains contingent and vulnerable.
Grid Dynamics’ heavy reliance on a concentrated customer base poses a significant and underappreciated risk to revenue stability, despite management’s emphasis on diversification. While the company highlights that its top 5 accounts are now entirely outside retail and include two global technology firms, a fintech leader, a U.S.-based global bank, and a leading financial institution, this concentration creates vulnerability. In Q1 FY26, revenues from the top 5 and top 10 customers reached 40.8% and 59.7%, respectively—up from 35.6% and 56.6% a year ago—indicating increasing, not decreasing, reliance on a small number of clients. The loss of any single top-tier account due to strategic shifts, budget cuts, or insourcing could disproportionately impact revenue. Furthermore, the company’s narrative around vendor consolidation benefiting Grid Dynamics assumes these clients will continue to expand their relationship, but there is no guarantee that consolidation will favor Grid Dynamics over other specialized AI vendors or that these enterprises won’t eventually bring critical AI workloads in-house as they build internal capabilities. The financial services and technology verticals, while currently strong, are also subject to intense competition from larger system integrators and niche AI specialists who may offer deeper domain expertise or broader platform ecosystems. Management’s confidence in being a “preferred vendor” is based on current engagement depth, but this status is not contractually locked and could erode if clients perceive better value elsewhere or if macroeconomic pressures trigger renewed cost-cutting initiatives. The market may be ignoring how fragile this preferred status is in an environment where enterprises are rapidly building internal AI centers of excellence and reevaluating external partnerships based on strict ROI thresholds.
The transition to AI-native, outcome-based, and fixed-price contracts introduces significant execution and financial risks that management has not adequately addressed, particularly regarding revenue recognition, margin volatility, and project overruns. While Eugene Steinberg highlighted the use of AI agents and the Rosetta framework to uncover requirement uncertainties during presale, and Anil Doradla acknowledged past learnings from fixed-price engagements pre-AI, the company admits it is still “experimenting” with monetization timing and revenue recognition for these new models. The shift from time-and-materials to fixed-bid work means revenue is tied to milestone achievement or outcome validation, not hours delivered—creating potential for revenue deferral, recognition delays, or even reversals if outcomes fall short of agreed criteria. Anil noted that the company is “taking baby steps” and that the optics of guidance look different from underlying business realities due to nonlinear timing of monetization. This uncertainty is compounded by the fact that Grid Dynamics is moving into higher-complexity domains like physical AI and mainframe modernization, where technical unknowns are more prevalent. For example, while the company successfully modernized a legacy mainframe platform for a home improvement retailer using AI agents, such projects involve high stakes—failure could result in significant reputational damage, financial penalties, or legal exposure if system integrity or regulatory compliance is compromised. The Rosetta framework and AI coding assistance may reduce uncertainty, but they do not eliminate it, especially in legacy systems with poor documentation or complex interdependencies. The market may be underestimating the likelihood of scope creep, technical surprises, or client disagreements over outcome metrics—particularly in early-stage AI deployments where success criteria are still evolving. If Grid Dynamics fails to consistently deliver on outcome-based promises, it could face margin compression, disputed invoices, damaged client trust, and a slower-than-expected ramp in non-T&M revenue contribution—undermining the very thesis that justifies its premium valuation.
Grid Dynamics’ operating model remains structurally challenged by persistent FX headwinds and rising cost pressures in its global delivery footprint, which are eroding margins despite growth in high-potential verticals. Anil Doradla explicitly cited FX fluctuations as a headwind of approximately $1.2 million on a year-over-year basis in Q1 FY26, impacting EBITDA, and attributed sequential and year-over-year EBITDA declines to a combination of FX headwinds and higher operating costs across delivery locations. The company’s non-U.S. headcount stands at 92.9% (4,611 employees), with significant exposure to Europe, Latin America, and India—regions where currency volatility and wage inflation are persistent. While the company utilizes natural hedges and an active hedging program, these are reactive measures that do not eliminate structural exposure. Furthermore, the shift toward AI-native delivery and productized engineering requires upskilling and retraining of talent, which increases labor costs in the short term. Although management cites internal AI automation gains (e.g., 2x recruitment productivity, 50% more RFP responses), these efficiencies are still nascent and may not yet offset the cost burden of retraining, geographic reorganization, and transaction-related expenses. The GAAP gross margin declined year-over-year from 36.8% to 34.8%, and non-GAAP gross profit margin fell from 37.4% to 35.3%, reflecting real pressure on core profitability. Management’s guidance for improved margins hinges on the success of new platform-based engagements, but if AI monetization lags or fixed-price projects underperform due to execution risks, the company could be caught in a squeeze: higher delivery costs from talent transformation without commensurate revenue uplift. The market may be overlooking how these structural cost pressures—exacerbated by geographic concentration in lower-wage but inflation-prone regions and currency volatility—could persist longer than anticipated, limiting margin expansion even as revenue grows in AI. Without clear evidence that internal AI automation is driving sustained, scalable cost reductions across delivery operations—not just support functions—the margin expansion story remains contingent and vulnerable.