Good afternoon and welcome to Globant's second quarter 2025 earnings conference call. I am Arturo Langa, Investor Relations Officer at Globant. All participants on this call will be in listen-only mode. After today's presentation, there will be an opportunity to ask questions. Please note, this event is being recorded and streamed live on YouTube. By now, you should have received a copy of the earnings release. If you have not, a copy is available on our website, investors.globant.com. We will begin with remarks by our Chief Executive Officer, Martín Migoya, our Chief Financial Officer, Juan Urthiague, and our Chief Technology Officer, Diego Tártara. This will be followed by a Q&A section. Before we begin, I would like to remind you that some of the comments on our call today may be deemed forward-looking statements. This includes our business and financial outlook and the answers to some of your questions. Such statements are subject to the risks and uncertainties as described in the company's earnings release and other filings with the SEC. Please note that we follow IFRS accounting rules in our financial statements. During our call today, we will report non-IFRS or adjusted measures, which is how we track performance internally and the easiest way to compare Globant to our peers in the industry. You will find a reconciliation of IFRS and non-IFRS measures at the end of the press release we published on our Investor Relations website, announcing this quarter's results. I will now turn the call over to Martín Migoya. Good afternoon, everyone, and thank you for joining us once again for Globant's quarterly earnings. It's always a pleasure to connect and share how we are continuing to evolve and improve. For us, innovation and reinvention are not occasional events. They are part of our DNA. Right now, we're building on that tradition with a clear and steady focus on the future. In the second quarter, we delivered revenue of $614.2 million, representing 4.5% year-over-year growth. Our pipeline is at an all-time high, $3.7 billion, up 25% from last year. Although the macro environment has extended sales cycles, our teams are laser-focused on converting this pipeline into signed work in the coming quarters, with large potential deals in healthcare, financial services, CPG, and gaming, among others. These large engagements, many already in advanced stages, position us well for conversion in the coming months, reinforcing our focus on high-value clients with strategic impact. Internally, we're fine-tuning the organization to be not only more efficient and profitable, but also more nimble and better aligned with the needs of the next generation of business models. Our aim is to ensure Globant is structurally agile and positioned to compete and win in an evolving landscape while continuing to deliver strong results for our shareholders. Juan will walk you through the details shortly. Globant's AI bots, which I introduced last quarter, are the virtual teams for the digital workforce. They're powered by Agentic AI and orchestrated by our experts. The way companies access our AI bots is unique in the industry through our monthly subscription model. It is a consumption-based, outcome-aligned pricing that provides guaranteed time and cost savings, shifting the value proposition to concrete results. After just one quarter, we already have 18 clients who have chosen this new model, and our subscription model accounts for a significant portion of our recent pipeline growth. We look forward to expanding this in the future. The AI world is moving at an incredible speed, and there are two major races going on. The first is over who builds the best foundational models or agents, and that race is OpenAI, Anthropic, Meta, xAI, and others. Every week, new models, frameworks, and tools emerge. This is not our race. The second is about who applies AI better, faster, and with greater return on investment on specific business cases for every company in every industry. For companies, the expansion of all these new foundational models presents an immense opportunity, but also a challenge of determining the right combination of models, tools, and approaches for their specific needs while managing data privacy and compliance. It is more like entering a dense and constantly changing forest where familiar paths disappear and new ones open overnight. In that environment, you need not just a map, but an experienced guide who knows how to adapt the route as the terrain shifts. At Globant, we take on that role, helping our clients find the safest and most effective path forward. We remain 100% client-centric while leveraging our deep partnerships with all major AI leaders to ensure our clients get the best of each ecosystem. This means we can select or combine the most advanced models and capabilities available in the market, integrating them in ways that are tailored to each unique task and business context. Globant Enterprise AI is the toolkit we bring into that complex forest. It is the golden path for our customers' generative AI adoption and impact. Offering seamless access to all major LLMs, it provides traceability, auditability, and granular access and cost control. It includes a library of hundreds of industry-tailored, plug-and-play business processes and a state-of-the-art Retrieval-Augmented Generation pipeline optimized for proprietary corporate data, delivering faster, more relevant, and contextually precise insights. Enterprise AI integrates with major enterprise platforms, manages AI-driven workflows across departments, connects human teams with intelligent agents, and is fully compatible with A2A protocols and MCP servers. The platform uses a token-based execution model to align AI usage to measurable business value. We'll launch the 2.0 version of Globant Enterprise AI. Diego will go into more detail on this shortly. When you combine that with the industry-specific expertise of our AI studios, you get a fully integrated engine for transformation across the entire AI value chain. We go beyond offering AI services. We architect and connect every layer of the AI stack, then deliver it as a scalable subscription-first solution. Our AI studios continue to drive deeper engagement with major Globant clients, unlocking cross-selling opportunities and deploying specialized talent from across our global network, unrestricted by regional boundaries. Part of growing is partnering with the best. A couple of weeks ago, we announced a multi-year collaboration with OpenAI as one of their few global services partners. By combining their world-class models with our engineering capabilities, we're delivering secure, responsible, and scalable AI adoption worldwide. We're already integrating GPT-5 across all layers of our Enterprise AI platform and embedding it into our AI bots processes. Along a similar line, days ago, we took a big step forward by becoming one of the few global partners to sign a strategic collaboration agreement with Amazon Web Services. Diego Tártara will expand on this later. We have teamed up with La Liga, Spain's premier football league, on a multi-year transformation program through Sportian, our sport tech joint venture. This is not just a technology project. It is a game-changer. By embedding our AI bots at the heart of their operations, we will unlock the full power of AI agents in sports to boost team performance, deepen fan engagement, and drive operational excellence across the league. It is a clear example of how Globant Enterprise AI is helping world-class organizations to embrace AI at scale and win in the most complex, high-profile arenas. We have strengthened our position in the robotics and AI ecosystem by investing in Inorbit, a leading robotics integration company. This expands our capabilities, enabling advanced orchestration of different fleets of robots and autonomous systems across industries. It ensures we can integrate physical automation into enterprise workflows at scale, connecting AI agents not only to digital processes but also to real-world robotic operations. Today, our revenue mix is more diverse than ever. North America remains our largest market with 54.1% of our revenue. Latin America accounts for 19.7% and is showing strong recovery with new records in bookings. Europe represents 19.6% and is our fastest-growing region sequentially, up 8.1%, with major wins in aviation and financial services. New markets grew an impressive 84% year-over-year and is currently 6.6% of our total revenue. The Middle East leads this surge, driven by our work on several giga projects. Our 100-square program continues to gain momentum with our client base growing and diversifying. Diego will share more on recent client wins and large-scale engagements that illustrate this strong momentum. 49 clients now generate more than $10 million in annual revenue, up from 39 a year ago. 339 clients generate over $1 million annually, up from 329 last year. Over the past 21 years, we have earned a unique position in our industry by fusing advanced technology, human creativity, and a deep understanding of our clients' needs. In 2025, the complexity of the environment is matched only by the scale of the opportunity as AI redefines business models, value chains, and competitive advantage. Our mandate is to help clients navigate this change with precision and foresight, capturing value across every layer of the AI stack. With AI bots, our AI-powered subscription model, AI studios, and the 100-square approach, we're delivering Globant as a full-stack AI company, one that designs, builds, and integrates technology platforms and industry-specific expertise into scalable solutions. This integrated model ensures we remain not just relevant but indispensable as enterprises embrace the AI-powered future. We're as energized by the challenges ahead as we are by the opportunity. With this team, I have no doubt we will capture them all. Thank you very much. Hello everyone. I'm happy to be back. I want to expand on the offering of Globant Enterprise AI 2.0 version that now includes MCP and agent-to-agent and the marketplace of agents for its users, among other new features. It enables a full AI adoption, addressing three core needs of our clients. First, governance. Organizations need to be in control of their own AI journey with strong security, risk management, traceability, and guardrails for both models and conversations. Globant Enterprise AI provides leaders with the control they need to map their exposure to both cost and risk. Second, build capability. Companies need the tools to create and deploy solutions connected to the enterprise systems. Globant Enterprise AI platform is home to the lab for agent creation, orchestration, and ensuring interoperability. As of this quarter, the lab now supports the Model Context Protocol and the Agent2Agent protocol. These enhancements allow seamless integration of agents and tools from across the AI ecosystem like Google Cloud, Azure AI Foundry, and Amazon Bedrock. Third, impact. The ability to explore, combine, and share solutions and to operate them so they can generate measurable business outcomes. This is where AI moves from pilot to scale adoption, ensuring that investments translate into tangible value across the organization. In this layer, we have recently launched a specific model for our clients called the Station. It offers a curated, searchable library of in-house built AI agents tailored to diverse business and industry needs, streamlining discovery while removing friction. With just a few clicks, users can now deploy agents via intuitive orchestration tools, speeding time to experimentation and value. Organizations using Globant Enterprise AI have reported an 80% reduction in legacy systems modernization times and a 50% increase in software development costs. Globant Enterprise AI also powers a lineup of hundreds of industry specifications, along with three flagship agents tailored to each of our core studios. Globant CODA is our Agentic suite for our digital studios, a key component of how we reimagine software development lifecycles and how we deliver value. Navigate is for our enterprise studio. It optimizes business operations and performance. Fusion from GUT Studio, launched at Cannes Lions this year, enhances full funnel marketing, communications, and advertising. It streamlines processes from content creation to campaign optimization. Now, let's discuss our work with some fascinating clients as we partner with them on their reinvention journey. In gaming, we're working with one of the leaders in producing real-time 3D content to deliver interactive solutions to high-growth sectors, including digital twins, automotive, healthcare, life sciences, and manufacturing. By employing our global delivery network, we will be helping this company to expand to new commercial markets, unlocking new business for them. On the technology side, we will be integrating their products into enterprise technology stacks, supporting new go-to-market strategies, and co-developing tailored solutions. Our Sustainable Business Studio is proud to be collaborating with the Worldwide Fund for Nature. Together, we're developing tools that link traceability with carbon footprint assessments. This initiative engaged multiple stakeholders, including national and other business entities, industry leaders, and civil society, facilitating the transition towards more sustainable meat production practices. The complexity and the time-sensitive nature of the challenges the WWF faced led them to enlist our Low-Code GeneXus platform. This collaborative effort will accelerate the adoption of sustainable practices, support regulatory compliance, and empower producers globally. We're excited to announce a partnership with a world leader in premium spirits to develop a generative AI-powered commercial insights agent. This innovative tool will provide our clients' employees with immediate access to critical data insights, streamlining decision-making in product development, marketing, sales, and strategy. By automating data retrieval, we're helping them to reduce the time and costs of traditional business intelligence workflows, allowing the teams to focus on strategic initiatives. The agent will enhance efficiency through self-service decision support and tailored recommendations. This initiative is just the beginning, as the commercial insights agents will lay the groundwork for future applications in brand planning, commercial forecasting, and innovation. We are redefining the potential of AI-powered enterprises and unlocking new growth opportunities. Regarding our partnerships, days ago, we advanced our relationship with AWS by establishing a strategic collaboration agreement to accelerate AI adoption and enable Globant to provide clients in specific industries with enhanced support for cloud migration, generative AI adoption, industry-specific solutions, while helping them to optimize their cloud usage and manage expenses efficiently. We're collaborating with Salesforce to deploy Agentforce and Data Cloud across multiple industries, automating their teams and enhancing their ability to support their customers, tailor marketing journeys, and better segment their clients. As Globant's creative industry pillar, GUT advanced on large-scale projects for top brands, including Progressive, Procter & Gamble, DoorDash, and more, as well as new projects for Havaianas, [Remoa], among others. Thank you, everyone, for joining us again. Hello, everyone. I will now review our Q2 2025 financial results before providing our outlook. Our performance this quarter is very aligned with our expectations back in May. Revenue for the second quarter came in at $614.2 million, representing 4.5% year-over-year growth or 1% in organic constant currency and 0.5% growth sequentially. Our non-IFRS adjusted operating margin was 15% for the quarter, holding steady despite some FX headwinds in Latin American currencies and demonstrating pricing and cost discipline in a tough market environment. Non-IFRS adjusted delivery EPS for the quarter was $1.53, an increase from the $1.51 we reported in the second quarter of 2024. Turning to the balance sheet, our cash and cash equivalents and short-term investments totaled $174.2 million. Net debt as of June 30 was $255 million. During this quarter, we increased our debt capacity to up to $1.1 billion. Free cash flow for the quarter was - $2.9 million compared to - $28 million from the same period last year. As always, we expect to generate strong free cash flow during the second half of the year. This quarter, we executed a business optimization plan. As explained by Martín, during Q2, we launched a new go-to-market strategy centered around our industry-specific studios and our 100-square accounts. The recently launched AI-powered subscription model, based on AI and our proprietary Globant Enterprise AI platform, is gaining traction with our customers. While we have delivered strong growth for many years, we have observed a more tempered demand environment over the last few quarters. The business optimization plan is part of our response to these organizational and demand changes and to best position ourselves for the next wave of growth. The primary goal of this plan is not only to protect our near-term profitability, but more importantly, to create the capacity to increase our investments in strategic growth areas for the rest of 2025 and beyond. This plan ensures we have the right talent and resources to execute on our AI-centric strategy and capture future opportunities while managing our cost base in the current market. The main actions under this plan included a comprehensive review of our workforce to align skills and size with our strategic priorities, which resulted in a reduction of approximately 1,000 employees or 3% of our workforce during Q2. A consolidation of our global office footprint based on an analysis of our facilities and lease contracts, and a strategic prioritization of our delivery centers to support future expansion. In connection with these actions, we recorded a one-time charge of $47.6 million in the second quarter. This plan should generate $80 million in annualized savings. These savings will be critical in protecting our profitability in the short term, despite the FX headwinds in Latin America, and will also be reinvested to fuel our growth engines, specifically our AI platform development and our people. We are taking decisive action now to build a more resilient and agile organization, ready to lead when the market accelerates. Now, let's talk about our business going forward. Based on current visibility for the third quarter of 2025, we expect revenue to be at least $615 million, which implies 0.1% year-over-year growth. This expected growth includes a positive FX impact of 50 basis points. We expect a non-IFRS adjusted operating margin to be at least 15%, and the IFRS effective income tax rate is expected to be in the 20%- 22% range. Non-IFRS adjusted diluted EPS is expected to be at least $1.53 per share, assuming an average of 45.6 million diluted shares outstanding during the third quarter. For the full year 2025, we now expect revenue to be at least $2.445 billion, representing 1.2% year-over-year growth. This expected growth includes a positive FX impact of 25 basis points. For the full year, we now expect our non-IFRS adjusted operating margin to be at least 15%, and the IFRS effective income tax rate is expected to be in the 20%- 22% range. Our full year non-IFRS adjusted diluted EPS is expected to be at least $6.12 per share, assuming 45.5 million diluted shares outstanding during 2025. Thank you for your continued support. Thank you, Juan, and hi, everyone. As we go through the Q&A section of this call, I will announce your name. At that point, please unmute your line and ask your question. Please mute your line after your question is done. I will also ask you to please limit your time to one question and to one follow-up only. With that in mind, thank you very much. We'll take the first question from the line of Tien-tsin Huang from JPMorgan. Tien-tsin, please go ahead. Great. Thank you, Arturo. Good to see everybody. I wanted to ask just on the AI-powered subscription model, and I think you mentioned on the subscription side, you had 10 clients that chose that model. I think I heard that. Can you tell us a little bit more about that? What work is being done, and how would the work and maybe the contract terms compare to what you would normally see in a more traditional model? Maybe start with that, if that's okay. Hi, Tien-tsin. Thank you so much for the question. Yes. It's actually 18 customers. 18, sorry. 18 paying customers. Yeah, look, the pipeline grew incredibly fast in terms of opportunities. We also were able to generate those 18 paying customers under the AI-powered subscription model, and that has been very well received by our customers. What we do is, on the back of all these customers, we're doing Agentic AI, generating the code, the development, the need, the software that our customers need. We are charging that with that subscription, in which we are taking the risk on our side of supervising what the AI agents create. Supervision, we hope that that supervision with time will go down. Now, it's at levels in which we want to be sure that we get the same quality as the traditional model that we have. We're extremely excited with what we are seeing with our customers and the type of contracts. We sometimes like the large portion of the discussion with the procurement offices and so on and so forth because it's kind of a new place where they never heard about. I'm very happy with the results. I'm very happy with the pipeline, how well received it was with our customers. They are used to or they understood the model of having a subscription and limits on tokens. I don't know, Diego, maybe you can. No, I think, Tien-tsin, for the most part, the most mature aspect is actually how we deliver value, which is how do we build software and the approach that we took. I think it was the most difficult and risky because this is actually changing, rethinking about the way of doing it. This is not infusing AI into an existing process. We reshuffle the whole thing to make the most out of it. That is actually complex because you need to convince your client that it has been working in a certain way for a long time. You even charge for this service in a different manner. You bring and make reality what has been a promise, which is the positive impact of AI on an enterprise environment. Like Martín said, reception was amazing. I thought getting traction out of this would have been a little bit more complicated, et cetera. The market and our clients understood this perfectly well. We received the right type of questions, which is amazing. The pipeline is growing healthy. I think I'm actually convinced that this is a major change that will definitely propel the future of Globant. By the way, Tien-tsin, the growth on the pipeline, a good portion of the growth of the pipeline between last quarter and this quarter is because of this. New conversations and new things are triggering up. We're very happy. No, thanks for doing that. I know it's a lot, and I respect that you guys are pushing for this so hard so quickly. It's great. You mentioned the pipeline, and it sounds like there's a lot of AI content overall in the AI side on the overall pipeline. Are you assuming a lot of conversion of the pipeline in the outlook in the second half? What's changed there? Do you expect some of these deals to close, for example, in the second half? Could they get pushed even further out given the newness and what's happening? Yeah. Yeah. We're seeing the conversion. I mean, the macro still is pretty uncertain, so we decided to go on the conservative side. We're seeing conversion in August and conversion like doing very well and much better than expected. We're positive about that. We want to remain very cautious around the idea of the second half of the year. Thank you so much, Tien-tsin, for your question. The next question comes from the line of Bryan Bergin from TD Cowen. Bryan, please go ahead. Hey, guys. Good afternoon. Thank you. I'll ask on the optimization. Maybe can you talk about how far through those initial changes you have progressed? We can see the 3% billable headcount reduction on a sequential basis. Should we expect any further activity kind of carry through into 3Q as well, or is that now through the system? Yeah. Hi, Bryan. How are you? On the headcount side, you're going to see some additional reductions happening in Q3, which already happened by now. The costs have all been accounted for during the second quarter because the plan was all provided at that point in time. There are also additional effects that are going to happen throughout the rest of the year in terms of the office consolidation and the talent development consolidation that we are doing. The vast majority of the plan has been already implemented, especially on the people side. You know, between what we did in Q2 plus some additional adjustments in Q3, that's already done. okay. My follow-up is on the creative performance. Can you give us a sense how GUT performed here, really the creative pillar relative to other studios? I ask because there's incremental concerns on the street about these particular areas from generative AI risks, right, and the ability for enterprises to do more themselves. I'm curious if you're seeing any trends specifically around that creative area that would support or refute that perspective. Yeah, it's a quite small operation, though, but it has been growing nicely. 100%. Yeah, we're very happy with that. I would say that from the quality and the caliber of work that we do, it's not the work that would be affected faster by the AI. I think that for those that are doing the outsourcing of generation of images and creation of specific campaigns for different channels, that impact could come faster. Our Fusion agent is targeting those kinds of customers that want to automate that pipeline of creation of content. I'm very bullish about the idea of how our entering into that creative space will grow in the future. What we are seeing is that between all the technology that is coming into the marketing space plus the amazing brand that we have with GUT in the space, we're creating very good momentum for the future. I'd say we are fully aligned with all the efficiencies that can be made in that space. I don't see that as a threat. Even more, I see it as an opportunity. When you look at how we create revenues in GUT, the vast majority comes from company branding, company positioning. It's very high-level, very strategic positioning of companies. It's not that much at the kind of campaign or short-term campaign or advertising that you could assume that might get impacted. We are not seeing impacts there. On the opposite, it's one of the fastest areas of growth at Globant. okay. Understood. Thank you. You're welcome. Thank you. Thanks. Thank you, Bryan. The next question comes from the line of Maggie Nolan from William Blair. Maggie, please go ahead. Hi. Thank you. I'm wondering if the Enterprise AI platform is creating enhanced stickiness with your customers compared to maybe traditional, more project-based engagements. That's a great question, Maggie. How are you? Look, I think that the Enterprise AI platform is like the golden path for generative AI adoption for our customers. It's an enterprise-class kind of integration of all the very, very complex AI ecosystem that is there to make it tangible, to make the things work. You don't just marry it with one LLM provider, but you can choose which to use. You can integrate all the workflows in your company, and you can connect with all the corporate information systems, and you can create your agents to generate those processes that companies need. We are using Enterprise AI for every single engagement on the AI bot side. We are using Enterprise AI for many customers that are finding or trying to find a safe path to implement AI inside their corporations. It's becoming a key component, as I mentioned on our last earnings call. It is a key component for the creation of the AI ecosystem inside corporations, where you don't just need to access LLMs, but you need to administer permissions, you need to administer access, you need to control costs of the things that you do. There's a lot of things that happen on the inner work of an enterprise-class implementation of AI that is brought to life by Enterprise AI. For me, it's extremely essential. It can be mounted on top of all the big hyperscalers' platforms. It can use many of those services. It is very well integrated into our solutions. I think moving forward, it will be a key component of everything we do. I don't know, Diego, if you want to. No, just to add to that, Maggie, I think Globant's type of services and delivery quality have created a great stickiness with our client. You can see that by the low churn and maturity, especially on the top accounts, which have more than 10 years with us right now. However, from a model perspective, in this specific case, you're actually the service as a software model where you actually build someone that's bespoke, but based on a platform, and you have the opportunity to operate that as well. I think it's a great way of outsourcing functions as opposed to the traditional BPO. From the software delivery perspective, you leave something with your client, which is they have the opportunity to use GeneXus Enterprise AI to build their agents, et cetera. We can help them build, customize, maintain, and operate solutions for them. From the core model itself, I think the answer is clearly yes. It provides more stickiness. Thank you. That's really interesting and kind of exciting to hear for the business model. One other thing I wanted to ask about the script. It sounded like there were a couple of larger deals that were close to closing. Maybe comment on those and the growth trajectory you're expecting by geography, by vertical kind of revenue growth, and how the pipeline is shaping up. Thank you. okay. Maggie, just a couple of deals that we are seeing now. Big deal on the financial service space, which is right in the final stages. Also, another big deal on the healthcare and life science space. Both of them in the U.S., which are pretty exciting to see how the market is recovering. We are seeing some deals on the enterprise side for CPG companies in Latin America. Latin America is recovering, as is explained on the numbers, too. We are seeing good signals across the board. It looks like the U.S. is recovering, Latin America is recovering, which was the two big things that we had on our last earnings call. I don't know, Juan, if you want to. Yeah, yeah, definitely. When you look at Latin America, after going down for a number of quarters, this is the first quarter it's sequentially up. I think that's good news. It shows like a stabilization. We are seeing and we are making progress towards the end of the rest of the year in Latin America. In the case of the U.S., you know, when we exclude some of the impact from professional services and some small impact on technology, the rest looks okay. BFSI with a very strong performance. Travel and hospitality, very strong performance. In general, yes, we are coming out of a number of quarters where the level of growth is lower. However, we start to see some positive numbers, especially on the pipeline. You look at the deals, it continues to expand. In a way, we are just getting ready for that recovery whenever it comes. Thank you very much, Maggie. The next question comes from the line of Jonathan Lee from Guggenheim. Jonathan, please go ahead. Great. Thanks for taking my question. Can you help unpack some of the assumptions around the revised growth outlook and maybe the level of conservatism you're assuming? How should we think about any potential acceleration off of your implied 4Q exit rate, if any? Yeah, sure. When you look at the guidance back in May and the new guidance, basically, there are a number of things that happened in the quarter. When we look at the second quarter, we were able to meet or slightly exceed, actually, the guidance that we provided. Looking into the full year number, as you remember, we had a significant reduction in the guidance for the year back in May. Now, you look at the EPS for the year, it is slightly up. There's a small tweak on the revenue line, mainly coming from one professional service customer and some small things that happen in technology. Overall, it's just a small tweak there. At the same time, when you look at Latin America, it's not coming down anymore. Sequentially, it was up this quarter. We are seeing kind of a stabilization and starting to build up based on the pipeline. When we look at the end of the year, I think that when we look at the U.S. economy, at some point, things will start to get better. We are not seeing further deterioration, which is good news, but things will have to get better. Companies will need to invest in the near future. They cannot withhold investments forever. That's something that's going to have to happen. When you look at the size of the pipeline, it continues to build. When you look at all the changes that we did in our go-to-market with the AI industry studios, the success that we are starting slowly but steadily showing progress on the subscription model, I think those are all good things that are shaping up. We expect that to start getting traction in the future. I think we are doing a lot of changes. We did a lot of changes on the structure. We did a lot of changes to have the right people with the right skills in front of customers, people with the right skills, given the new technologies that are coming that resulted in a significant business optimization plan. We're doing lots of things to make that progress that I think is going to happen in the near future. Yeah, also, as a general comment, I think that the amount of the opportunities are growing is showing up on the pipeline. I would say that as a general understanding of all the things that are happening is that the amount of new projects, as we have been saying for many, many, many quarters, the amount of new projects is incredible. The complexity in any corporate environment to implement any of these agents into production with the right railways for the LLMs, with the right access permission, with the complexity of connecting with production data, with the complexity of implementing that and then taking it and spreading it out in a big organization, we're living it ourselves. We want to implement agents. It takes us a while to generate, even using the most sophisticated tools, it takes us a while to make it happen in the whole organization because we need to connect to complex systems. That is creating a massive set of opportunities that before didn't exist. This is on top of all the digital transformation work. This is on top of all the enterprise migration that is happening and will keep on happening. I think the opportunity for us is massive. It's reflected on actual numbers. Post-conversion has been slower, but I believe that little by little, these corporations will come to us saying, listen, we need help to implement this. Now we have seen the first cases. You know, a year ago, I was saying this is like very small and proof of concepts that are happening. Now we're seeing customers say, no, I want a full AI transformation program to change my processes because I discovered that it's not that easy to make it happen. We need help. That's creating a massive opportunity for us. I believe that's why I'm so bullish about the future of our company and all the moves in which, of course, new ways of delivering the same things are needed. We are delivering that. I'm extremely excited about what's coming. Thanks for that, Carolina. On the heels of those opportunities, can you talk through some of the pricing discussions you're having, particularly around potential impact to pricing from your AI-powered subscription model, as well as just given the rather competitive environment that you're seeing today? Yeah, Jonathan, listen. The pricing of our AI bots leaves us with much better margin than what we have on a traditional project. I think that explains by itself. I don't know, Juan. At the same time, it has better margins. It is cost-effective for the customer as well. It's a model that so far, the way we are selling it, the way we are contracting with customers is a win-win. They are getting part of the productivity gains. We're getting part of the productivity gain. It's working well. We need to make sure, as Martín said at the beginning, that we continue to deliver the same quality that we have always delivered to our customers. Over time, as the coding gets better, the agents get better, we should even be able to hopefully improve a little bit those margins. It's marginally positive. At pricing, it gives customers some efficiencies straight away. Thank you very much. The next question comes from the line of Divya Goyal from Scotiabank. Divya, please go ahead. We cannot hear you, Divya. It looks like you're on mute. No. Let's go to the next question, and we'll come back to Divya. I'm sorry. The next question comes from the lines of Sean Kennedy from Mizuho. Sean, please go ahead. Hi, everyone. Thanks for taking my question. I was wondering about the North American deceleration. Was that due to certain planned projects ramping down and lower conversion the last few quarters? How has the North American pipeline conversion been trending since May? Thank you. Thank you, Sean, for the question. North America was sequentially down 2%. It's very much focused on a customer in professional services and some small customers in technology. When you look at the pipeline, when you look at some of the deals that we recently closed that Martín mentioned in healthcare, in PFSI, we are optimistic. I think that after several quarters of instability and changes and tariff discussions and macro and all that, things are not great, but at least they are more stable in a way. We look at the pipeline. The majority of the pipeline is getting built out of the U.S., and we believe that with the new go-to-market with AI studios, industry studios, we're going to be able to take advantage of that. We are positive on North America. Of course, Martín mentioned at the very beginning, deals take longer to close. That is something that was not common in the past. At least a few years ago, we are now living with that. We start to see some big deals getting closed, which was difficult a couple of months and quarters ago. Bigger deals are starting to close again. Got it. In Latin America, it was nice to see the sequential growth there. Could you kind of discuss some more detail about what you're seeing on a country-by-country basis? Argentina is probably the best-performing country around this time in the region. We start to see recovery in Brazil, Peru, and Mexico. They are not all of them growing, but we are not seeing the deterioration that we saw, for example, in Brazil and Mexico earlier this year or late last year. We see the stabilization there, some recovery in the case of Mexico, and a very strong Argentina. Great. It's great to hear. Thanks for all the color. Appreciate it. Thank you. Thank you. Thank you, Sean. The next question comes from the line of Nate Svensson from Deutsche Bank. Nate, sorry, please go ahead. Yeah, thanks for this, guys. I wanted to follow up on Jonathan's question on pricing. Nice to hear that the new subscription-based model is going to be accretive. If we think about sort of revenue per head dynamics, excluding that, you know, just 18 projects so far, so a relatively small portion of the base. I think last quarter revenue per head was up 2.8%. We have it a touch lower, maybe closer to 1% this quarter. Just wondering how you're thinking about pricing trends for the overall book of business for the remainder of the year, especially given all the dynamism and uncertainty in the macro. Yeah. So far, we have been able to either maintain or slightly grow our revenue per head, which I think, given the market that we have seen in the last two to three years, is quite remarkable. We have been, maybe at the expense of sometimes growth, very consistent on trying to maintain our pricing, trying to maintain our margins, working hard to protect profitability. Somehow, the results have been showing that over the last two years. Pricing, yes, it's not a very hot market. Pricing is a challenge, and you have to negotiate. As a company, we believe that we need to protect profitability. We need to have deals that make sense for the company, right? We don't want to, you know, we always try to protect margins since we are public. Our margins, actually, if you look at gross margins, they've been between 38% and 41% for the last 10 years, 12 years as a public company, right? For us, we know it's easy to take prices down, but we work very hard to avoid those situations. The revenue per head shows that as well. Yeah, it makes sense. It's good color. I guess related to margins, I wanted to ask about utilization as well. I think last quarter on the call, you talked about improving utilization as a potential margin lever opportunity. I think last quarter, utilization fell sequentially and year over year to call it 78%. Any update on what that was this quarter? In light of the business optimization, I guess my guess would be that utilization maybe gets a little better, but expectations for the rest of the year. Yeah, yeah. Utilization is up about 40 basis points this quarter, and we expect to take it even further up. You know, with all the business optimization plan that we did, part of that creates or improves our utilization rates as well. We always talked about 80%- 81% as a target utilization number, and that's still valid. We're working towards that. We also mentioned in the past that, you know, in a very hot market, it's always easier to drive utilization up. I mean, we are making progress, and this quarter, we were able to increase that by about 40 basis points. Thanks for that, Carolina. Thank you. Thank you. Thank you, Nate. The next question comes from the line of Leonardo Olmos from UBS. Leonardo, please go ahead. Hi, everyone. Thank you for taking the question. Good evening. Can you talk a little bit about the potential or not concentration of consultancy firms or IT services among clients? I mean, are clients talking to newer firms, or are they just concentrating on the biggest one? I thought, and please correct me if I'm wrong, that you may be not growing that much with clients with revenue around on top of $1 million. If that's the case, please correct me. All in, I want to know how competition is and how prices are and if clients are concentrating with a lower number of IT services companies. Thank you. Yeah, we have been running a lot of these consolidation processes. The good news is that we have been selected in all of them, or in pretty much all of them. That's one thing. The second thing is we have more accounts yielding more than $1 million than a year ago. That is fixed by itself to your question. The other thing is, yeah, companies are looking for a differential offering. They're tired about the traditional, you know, big consult, massive consulting firms that do not innovate or that they are not bringing anything new to the table. One of the reasons why we're being able to maintain our position and leadership is that we are offering something new. We are telling to the market, listen, the model will change. We are changing faster than that. I'm extremely excited about that. I think that the future will keep on showing that smaller companies like Globant have a much higher impact on how corporations transform themselves to become the next versions of themselves. I'm very bullish on that. Please, my teammates here. I think, like Martín said, when the consolidation happens, you know, cyclically, on a cycle. We've been through many of them. Most of our top clients have done that. Those were actually opportunities for us because we ended up, you know, in the vast majority, we ended up selected. I think that at this particular moment, what's interesting is that, you know, procurement, instead of coming up with consolidation, et cetera, because of a requirement of the business, it, on the contrary, opens up. What this means is it opens the possibility to bring boutique type of compacts. Why is that? They're smaller. They react fast to changes in the technological landscape. Typically, when these major changes happen, you see the opposite. As opposed to consolidation, you know, open it up and soften the vendor selection. What's good about this is that we've been running alongside those companies. We've been getting those types of opportunities. In most of our clients, we're either leading AI or executing our client's strategy. Many of our clients have either selected a Chief AI Officer or their Chief Data Officer has put together a strategy. They say, you know, Globant, now I need your help to scale this. I think that's the most important aspect that I would like to quote. The other process has been running over and over for a long time. When this happens, you set apart companies that innovate fast, that take bold decisions from the ones that take longer. Yeah, also, our focus on the 100-square program, right? I mean, we're paying a lot of attention to those accounts that can multiply for us, that have the potential of having a larger need of making these massive transformations. I think that approach is extremely important for our future. We'll keep on paying a lot of attention and our best people into those accounts. Crystal clear. Thank you very much, Martín and Diego. Sure. Thank you. Thank you very much, Leo. That will be all for the Q&A section of today. Thank you all. With that, I will now turn it over to Martín. Let's try to. You want to try, Divya? Divya? Divya? Divya works. Hi. Can you hear me? Can you me? Can you all hear me? Yes, I hear you, Divya. Thanks for bringing me back. I really appreciate it here. I just wanted to give it a little bit more color. I know you've provided a lot of information on the AI pods. Could you help us understand what's the scale and scope of some of the engagements you're seeing? Are you seeing some of these engagements expand into existing clients? Are you also seeing some net new clients adopting and getting interested in this AI pods methodology or solution offering that you have? Divya, listen, the pipeline contains a lot of new customers. I would say more than half are current customers. The conversion itself has been on current customers and the new customers and some new customers. I would say that the vast majority of what we close today are current customers plus two or three new. It's very interesting to see the dynamics inside the pipeline because inside the pipeline, you have a lot of customers transforming deals and making deals that used to be like traditional deals, but now moving into this new AI-powered subscription model. We have a lot of new customers that we didn't have before. We announced it. It was quite low profile. We just sent emails to some of our customers. I received, for example, an email from a hyperscaler saying, hey, Martín, look, we understand this quite well. We like what you're doing. We would love to know more about it. That happened yesterday. What I'm saying is it's creating momentum. Bain Consulting put out a report to talk about the new model. It was a very interesting report talking about how this new model should be the next choice model. We're very excited. I don't know. I think you've covered it all. We've covered it all. okay, that's the answer, short answer. You have one more. It's great. It's very exciting. I can tell you, as an outcome-based pricing model picks up across the sector, you're definitely leading the trend, is what I would say. One more question that I wanted to understand. I know some of your peers have been talking about helping global enterprises set up global capability centers, so GCCs. Is that something Globant has been participating in, helping the global enterprises grow their offshoring unit? What is the role, if at all, that you play in there? That's all for me. Yes. The answer is yes. We are participating in many of those deals. One of the deals that we're just about to close or is closed already is about taking over one of these centers. Our proposal, as opposed to the traditional proposal, is a proposal that includes our AI pods. It's something that we're pretty proud of because we won because of that. I think that is something that, I mean, this new strategy, this new AI pods model, comes with two things. First, it makes savings tangible. Second, it provides a level of transparency of what we do that is unparalleled to any other model in the past. We can give you a report of every single token that we use and how we use it, and which is the artifact that has been created using those tokens and how we supervise that and how we make sure that all those things make sense. In essence, I believe that it's a breakthrough on the offering. It's something that we're extremely excited about. Yes, we're participating in all these deals. Hopefully, we will win them all. You're the best. Thank you so much. Thank you very much. Thank you, Divya, and I apologize for that. Thank you. With that, we conclude the Q&A section for today. Thank you all. I will turn it over to Martín for some closing remarks. Please go ahead, Martín. Thank you, Arturo. Thank you, everyone, for participating, for your continued support. Looking forward to seeing you on our next earnings call. Ciao. Bye-bye.