Good day and welcome to the Second Quarter 2024 EPAM Systems Earnings Conference Call. All lines have been placed on mute to prevent background noise. After the speakers' remarks, there will be a question-and-answer session. I want to inform all participants that this call is being recorded. Thank you. I will now hand it over to Mike Rowshandel, Head of Investor Relations, to start the conference. Mike, go ahead.

Good morning, everyone and thank you for joining us today. As the operator just mentioned, I'm Mike Rowshandel, Head of Investor Relations. By now, you should have received your copy of the earnings release for the company's second quarter 2024 results. If you have not, a copy is available on epam.com in the Investors section. With me on today's call are Arkadiy Dobkin, CEO and President; and Jason Peterson, Chief Financial Officer. I would like to remind those listening that some of the comments made on today's call may contain forward-looking statements. These statements are subject to risks and uncertainties as described in the company's earnings release and SEC filings. Additionally, all references to reported results that are non-GAAP measures have been reconciled to the comparable GAAP measures and are available in our quarterly earnings materials located in the Investors section of our website. With that said, I will now turn the call over to Ark.

Thank you, Mike. Good morning, everyone. Thank you for joining us today. First, I would like to start off with our second quarter results which came generally in line with our expectations. We believe our performance in the second quarter of 2024 reflects our continued strong execution and adaptability amidst a still complex demand environment. Let me share some current highlights of our business from Q2 up to today. Our underlying business continues to show signs of stabilization. In the second quarter, we delivered very strong growth in our healthcare and life sciences vertical and strong growth in our emerging verticals, where we also saw some slight sequential improvements in financial services. In some of the verticals, namely business information and media, we continue to work through the impact of the ramp-downs from the few large clients we have mentioned before. On the demand environment, we do see broad-based signs of stabilization as well across both EMEA and North America. At the same time, clients are still cautious with larger programs and our visibility to a significant increase continues to be constrained by a mix of clients, cost-saving priorities, delays in program starts and clients' own business changes. As a result of this complex environment, we are currently assuming no net improvements in overall demand for the remainder of the year. Jason will provide more details of our updated outlook for 2024. To be clear, notwithstanding the overall demand picture, we are optimistic about certain sectors of our target market and our current portfolio returning to a modest growth story in the next two to three quarters as we see it now. Overall, while we adjust our offerings and our delivery mix to the parameters of the current demand environment, we continue to see significant traction in our data and analytics, core engineering and digital engagement offerings, especially through the broad transformation of reach with AI, as well as increasing evidence that our consulting and advanced technology capabilities are driving new top-of-the-funnel opportunities for us. Meanwhile, we remain focused on responsibly managing our business in this part of the cycle, building on our strong fundamentals and ensuring that EPAM continues to be the partner of choice while the demand environment improves. Turning now to our global delivery strategy. In Europe, our differentiated capabilities continue to create significant opportunities for our clients to leverage top talent on their most complex business and technical challenges. We believe this traditional EPAM advantage will continue to serve our clients well, especially as they turn from mostly cost-driven considerations back towards driving growth and innovation through the use of advanced technologies for their complex transformation and modernization efforts. We are continuously investing in our more recently established delivery hubs across Western and Central Asia which are emerging talent markets and where we are enabling strong and experienced tech talent to responsibly lead our future growth in the region. India remains a priority and is on track to become our largest delivery location by the end of the year. We have built scaled centers of excellence in data and analytics, digital marketing, commerce, salesforce, SAP, and other key horizontal and vertical service lines. Our primary focus is on building a scaled EPAM-grade quality engineering, while blending in with many unique enterprise-level capabilities present in India and not available in most of our other EPAM locations around the globe. We believe this approach will differentiate EPAM India from our clients and create strong growth opportunities for our people. We are also continuing to expand our core engineering capabilities in LatAm. In addition to Mexico and Colombia, we now have a delivery hub in Argentina. We will continue assessing and developing new local talents across the region. In each of these locations, we are investing in our existing and new technical capabilities, including crucial-for-the-future GenAI data, machine learning, and predictive AI along with corresponding IP development. In short, we are building full-service GenAI delivery practices through a network of GenAI Labs across major development centers to enable and scale GenAI-enabled client production activities. In addition, both in India and LatAm, we are evolving strategies that include not only differentiated delivery capabilities but also being able to offer compelling in-market presence end solutions, particularly as we seek to serve our global alliance in a more complete and strategic manner, which means building locally a much stronger partner ecosystem as well. Finally, in all our locations and specifically in our major client markets, we continue to focus on our client engagement programs and to improve our consulting industry capabilities across our service offerings. Now let's turn into a bit more detail on our GenAI approach. For the last decade and despite all challenges during the last several years, we have continuously been recognized as a leader in advanced technology, digital engineering, and complete data transformation programs. This naturally extends to EPAM being regarded as the company that understands the complexities of AI transformation, something we have been doing for ourselves, our partners, and our clients for some time now. Today, I would like to highlight our up-to-date progress in that area and how EPAM is helping clients pragmatically initiate and then move use cases beyond pilots into production deployments. Our current approach to AI transformation is three-dimensional. Dimension one: EPAM internal transformation and GenAI enablement investments. We set an ambitious goal for ourselves to upskill and effectively train an absolute majority of the company on the usage of GenAI fundamentals and to do so both responsibly and with the EPAM-level technical depth. A dedicated program was established to execute this, and today, with the help of our educational platforms, internal specialized tools, and our global maintenance community, close to 100% of the EPAMers have gone through training and are applying AI in their daily work activities. While most companies have announced similar programs, we believe that during the last 24 months, our early and highly focused efforts across a broad range of EPAMers allow us to better understand future opportunities and to invest in differentiated intellectual property and accelerators around GenAI-enabled engineering solutions. Our combination of training, intellectual property, and open-source style internal initiatives have now become drivers of scale in our advanced GenAI practitioner communities across all EPAM organizational units and practices. We assume that well over 10% of EPAMers are now advanced GenAI technical practitioners, while over 1,000 are becoming strong internal AI champions with the ability to lead GenAI-enabled business solutions. We believe all that has enabled our dimension two: client transformation opportunities. Our AI client projects today have evolved from exploratory pilots and proof-of-concept late last year to now EPAM being selected by clients as a primary AI partner with involvement in hundreds of GenAI-led engagements. We are helping to change the full value chain of SDLC from one side and enabling the implementation of real GenAI-driven business use cases from another. Let me briefly highlight just a few intellectual property investments. EPAM DIAL is a unified GenAI orchestration platform helping enterprises spin the experimentation and innovation efforts to implement real business use cases and GenAI-enabled solutions by connecting into meaningful workflows and load balancing a set of public and proprietary LLMs and SLMs together with different types of internal and external data sources, AI-native applications and customer dots. EPAM AI/Run is a GenAI-powered delivery framework that accelerates the entire software development life cycle and helps clients recognize ROI of their AI investments by improving time-to-market speed by up to 30%. EPAM EliteA is a comprehensive collaboration platform for teams that streamline the development, accessibility, and management of large language model assets, including prompts, templates, and agents. Now a few specific examples for clients who operate across completely different user environments. For Unity, the world-leading platform of gaming tools for creators to build and grow real-time games, apps, and experiences across multiple platforms, we helped build Unity Muse. We supported multi-cloud migration to Microsoft Azure to aid in GenAI capabilities to make game creation faster and easier for one million developers by using natural language prompts to generate sprites, textures, and animations. We also provided chat-based assistance and troubleshooting, as well as the ability to even create behavior trees. For XSOLIS, a leader in healthcare system purpose-built solutions and industry-leading AI, we helped to develop a new generation AI platform on AWS enhancing Dragonfly, XSOLIS's flagship product. The platform provides real-time data and predictive analytics to nurses, case managers, physician advisers, utilization management, and revenue cycle leaders across 500 hospitals and health systems with more than 500 payer connections. Finally, for the IMF, as part of modernizing their data platform, we built StatGPT which is an SDMX-driven GenAI application for statistical organizations, allowing their users, economists, and statisticians to truly transform, analyze, visualize, and interpret statistical data using a natural language interface via proprietary talk-to-you-data framework powered by EPAM DIAL and EPAM Quanthub accelerators. Initial results showed a 50% increase in research productivity and a 35% increase in research accuracy. Overall, we are seeing a significant rise of GenAI-led engagements across every vertical and a very broad set of use cases. This is now driving transformation in both front-end customer experiences as well as significant back-office and process-related use cases. Finally, all that in turn allows us to enable dimension three for AI transformation: extended client network, first of all, because we saw that we are very practical in our approach. Our technology and AI transformation programs are much larger and more complex today than where we started just a few quarters ago, encompassing both consulting and engineering services as well as a broader range of partnerships. With more than 150 strategic partners, we're accelerating modernization, adopting cloud-native architecture and leveraging AI and advanced analytics, particularly when our clients' projects have a high degree of technical and business complexity. We are our partners' choice for making corporate engagement real. Recently, we've been named Partner of the Year by several of our cloud data and digital partners, including Databricks' Disruption Partner of the Year, recognized for the industry-leading design and implementation of GenAI-powered conversational interfaces, state-of-the-art machine learning, and large language model framework. As an elite level partner, we are one of only five companies listed in Databricks' center of excellence. We are also recognized as Google Cloud Talent Development Partner of the Year for our commitment to training, upskilling, and reskilling our team in cloud and AI skills. Additionally, we've received the Microsoft Gaming Partner of the Year award for pushing the boundaries of creativity and technology. Lastly, we were recognized by Mark in commerce tools for delivering the best-in-class modern commerce experience for our clients. It's easy to assume the next question: what is the revenue impact of these programs? I guess the answer is probably predictable as well. We are still in the early days of the AI wave. But at the same time, we are very optimistic about the accelerating pipeline opportunities coming from AI-led transformation and what that can bring downstream for us as a highly trusted and valued partner. We remain focused on expanding our efforts to drive demand, remaining relevant to our clients and what they need, and proactively expanding our global market share. We also remain committed to managing our delivery footprint, expanding to cost-efficient locations and generally optimizing our ways of working, so we can continue to provide premium service at an attractive value to our global client base. I firmly believe EPAM continues to be well positioned to capture a rebound in market demand, driven by long-term pressures for legacy modernization with needs for advanced customer-centric solutions and by significant interest in how to apply GenAI and GenAI capabilities to build new platforms and solutions. With this, I would like to pass to Jason to provide additional details on our results in Q2 and our future performance.

Thank you, Ark and good morning, everyone. In the second quarter, EPAM generated revenue of $1.147 billion, a year-over-year decrease of 2% on a reported basis or 1.7% in constant currency terms, reflecting a negative foreign exchange impact of 30 basis points. Due to our exit from the Russian market, we no longer generate revenue from Russian clients. The impact of this exit had an approximate 50 basis point negative impact on year-over-year revenue growth. Excluding Russia revenues, year-over-year revenue for reported and constant currency would have decreased by 1.5% and 1.2%, respectively. Moving to our vertical performance, life sciences & healthcare delivered very strong year-over-year growth of 22.4%. Growth in the quarter was driven by clients in both life sciences & healthcare. Consumer goods, retail, and travel decreased 7.7% on a year-over-year basis, largely due to the declines in retail, partially offset by solid growth in travel. Financial services decreased 5.6% year-over-year, driven by softness in asset management, banking, and payments. In the quarter, the vertical delivered slight sequential growth, indicating stabilizing demand. Software and hi-tech contracted 3.7% year-over-year. Business information and media declined 12.6% compared to Q2 2023. Revenue in the quarter was substantially impacted by the previously discussed ramp down of a top 20 client. Finally, our emerging verticals delivered solid year-over-year growth of 10.6%, driven by clients in energy and telecom. From a geographic perspective, the Americas, our largest region, representing 60% of our Q2 revenues, grew 1.8% year-over-year on a reported basis and 2% in constant currency terms. EMEA represented 38% of our Q2 revenues, contracting 6% year-over-year and 5.6% in constant currency. APAC declined 0.6% year-over-year or 0.2% in constant currency terms and now represents 2% of our revenues. In Q2, revenues from our Top 20 clients declined 3.7% year-over-year, while revenues from clients outside our top 20 declined 1.1%. The relatively stronger performance of this latter group was driven by both new client and inorganic revenue contributions. Moving down the income statement, our GAAP gross margin for the quarter was 29.3% compared to 30.9% in Q2 of last year. Non-GAAP gross margin for the quarter was 30.8% compared to 32.6% for the same quarter last year. Relative to Q2 2023, gross margin in Q2 2024 was negatively impacted by the strengthening of currencies associated with certain delivery locations, as well as the impact of compensation increases, including those resulting from our recent promotion campaign which we were not able to offset through pricing. The negative impact of foreign exchange and compensation increases exceeded the benefit of improved utilization. GAAP SG&A was 16.9% of revenue compared to 16.7% in Q2 of last year. Non-GAAP SG&A in Q2 2024 came in at 14.3% of revenue compared to 14.8% in the same period last year. SG&A improvement in the quarter is the result of our ongoing focus on managing our cost base and increased efficiency in our spend. GAAP income from operations was $121 million or 10.5% of revenue in the quarter compared to $144 million or 12.3% of revenue in Q2 of last year. Non-GAAP income from operations was $175 million or 15.2% of revenue in the quarter compared to $191 million or 16.3% of revenue in Q2 of last year. Our GAAP effective tax rate for the quarter came in at 26.3% and our non-GAAP effective tax rate was 24.3%. Diluted earnings per share on a GAAP basis was $1.70. Our non-GAAP diluted EPS was $2.45 compared to $2.64 in Q2 of last year, reflecting a $0.19 decrease year-over-year. EPS was positively impacted by a Serbian government investment incentive received and recognized in the quarter which improved Q2 diluted EPS by $0.06. Although the benefit from this incentive was contemplated in the full-year guidance communicated during our Q1 earnings call, at that time, we had expected to receive the incentive and recognize the benefit in Q3. In Q2, there were approximately 58.1 million diluted shares outstanding. Turning to our cash flow and balance sheet, cash flow from operations for Q2 was $57 million compared to $89 million in the same quarter of 2023. Free cash flow was $52 million compared to free cash flow of $82 million in the same quarter last year. At the end of Q2, DSO was 76 days and compares to 73 days for Q1 2024 and 71 days for the same quarter last year. The uptick in DSO reflects an increase in the time some clients are taking in the review and approval of payments, combined with the last few days of the quarter falling on a weekend. Share repurchases in the second quarter were approximately 1.16 million shares for $214 million at an average price of $184.97 per share. In June 2024, EPAM completed the share repurchase program authorized on February 13, 2023. Over 16 months, 2.24 million shares were repurchased at an average share price of $222.90. On August 1, 2024, the Board of Directors approved a new share repurchase program with authorization to purchase up to another $500 million of EPAM common stock over a term of 24 months. We ended the quarter with approximately $1.8 billion in cash and cash equivalents. Moving on to a few operational metrics, we ended Q2 with more than 47,000 consultants, designers, engineers, and architects, a decline of 4.8% compared to Q2 2023. Sequentially, production headcount remained unchanged as the company reduced headcount in certain on-site locations, while continuing to hire in India. Our total headcount for the quarter was more than 52,650 employees. Utilization was 77.5% compared to 75.1% in Q2 of last year and 76.8% in Q1 2024. However, on-site utilization remains below targeted levels, and the company will continue to take actions to optimize on-site resource levels to improve utilization. Now let's turn to our business outlook. Although client demand has stabilized, we continue to see very little improvement in the near-term demand environment. We are experiencing growth in certain verticals, seeing relatively high levels of new logo activity, and working with clients to bring GenAI programs into production. We are also beginning to see some constructive improvement in client discussions regarding future programs. However, decision-making continues to be relatively cautious as some clients continue to face challenges with their own end markets and revenue generated by individual new logo accounts is, on average, less than that generated in prior years. Although we believe clients are beginning to more actively engage around new initiatives, our guidance assumes macroeconomic stability with no improvement in the aggregate demand environment for the remainder of the year. We are hopeful that a change in the tone of client conversations will result in an improved demand environment in 2025. For the remainder of 2024, we are expecting a slight increase in Q3 revenue relative to Q2 driven by greater build-outs in the quarter, substantially offset by higher vacation levels. We are expecting a modest sequential decline in Q4 revenues, driven largely by some of the seasonal factors mentioned previously. We are maintaining our focus on demand generation and will continue to prioritize revenue growth for the remainder of 2024. At the same time, we are taking steps to improve cost efficiency and on-site utilization, and now expect to operate at a higher level of profitability in the fiscal year. Finally, our operations in Ukraine continue to run at high levels of utilization, a testament to our team's dedication and focus on maintaining uninterrupted quality of delivery. Our guidance assumes that we will continue to be able to deliver from our Ukraine delivery centers at the productivity levels similar to levels achieved in 2023. Moving to our full year outlook. Revenue is now expected to be in the range of $4.590 to $4.625 billion, a negative growth rate of 1.8% at the midpoint of the range. The impact of foreign exchange rate growth is expected to have a positive impact of approximately 10 basis points. At this time, we expect approximately 1% of revenue contribution from already completed acquisitions. We expect GAAP income from operations to now be in the range of 10.5% to 11% and non-GAAP income from operations to now be in the range of 15.5% to 16%. We expect our GAAP effective tax rate to now be 21%. Our non-GAAP effective tax rate, which excludes excess tax benefits related to stock-based compensation, will continue to be 24%. For earnings per share, we expect the GAAP diluted EPS will now be in the range of $7.18 to $7.38 for the full year; and non-GAAP diluted EPS will now be in the range of $10.20 to $10.40 for the full year. We now expect a weighted average share count of 57.9 million fully diluted shares outstanding. Moving to our Q3 2024 outlook. We expect revenue to be in the range of $1.145 billion to $1.155 billion, producing a year-over-year decline of 0.2% at the midpoint of the range. On a constant currency basis, we expect Q3 revenue to be flat year-over-year. For the third quarter, we expect GAAP income from operations to be in the range of 10% to 11% and non-GAAP income from operations to be in the range of 16% to 17%. We expect our GAAP effective tax rate to be approximately 24% and our non-GAAP effective tax rate to be approximately 24%. For earnings per share, we expect GAAP diluted EPS to be in the range of $1.75 to $1.83 for the quarter and non-GAAP diluted EPS to be in the range of $2.65 to $2.73 for the quarter. We expect a weighted average share count of 57.4 million diluted shares outstanding. Finally, a few key assumptions that support our GAAP to non-GAAP measurements for the remainder of the year. Stock-based compensation expense is expected to be approximately $45 million for each of the next two quarters. Amortization of intangibles is expected to be approximately $6 million for each of the remaining quarters. The impact of foreign exchange is expected to be a $1 million loss for each of the remaining quarters. Tax effective non-GAAP adjustments are expected to be around $13 million for each of the remaining quarters. We expect excess tax benefits to be around $1 million for each of the remaining quarters. Severance driven by our cost optimization program is expected to be around $10 million for each of the remaining quarters. With our significant cash position, we are generating a healthy level of interest income and now expecting interest and other income to be approximately $13 million for each of the remaining quarters. While we work our way through this cycle of lower demand, we will continue to run EPAM efficiently, positioning the company to capitalize on a more normalized demand environment. Lastly, my continued thanks to all our employees for their dedication and focus on serving our clients and driving results for EPAM. Operator, let's open the call up for questions.

Your first question comes from the line of Maggie Nolan of William Blair.

I wanted to dig into the utilization and the dynamics between on-site and offshore. So first of all, what percentage of the workforce is considered to be on-site this quarter? And then is offshore utilization running higher than what you view as sustainable to offset some of that weakness on on-site because you're not too far off from your historical range here?

Yes, we believe that offshore utilization is quite strong. I'm currently having difficulty recalling our exact on-site percentage, but in terms of total headcount, our utilization is definitely lower than our typical levels. This remains an area where we face some challenges, and it may also be affecting our revenue growth for the remainder of the year, as we are seeing increased demand for offshore services and more demand from India. We are still operating at slightly lower on-site utilization levels, and we are actively working to improve this through demand generation efforts and other actions I mentioned in my prepared remarks.

Got it. And then somewhat related, just building on that, the improvement in the margin outlook, is that primarily related to those actions that you want to take on utilization? Are there other levers you're pulling? And have you already seen some progress here in the third quarter to fuel that optimism in the increased number that you gave?

No, absolutely. The focus has been on cost optimization after we reset our expectations for revenue growth. So we've been more efficient in corporate functions and SG&A. We've been working on utilization. The actions that we intended to take are underway, and it is beginning to show up probably even in a little bit of the benefit that we saw in profitability in Q2.

Your next question comes from the line of Bryan Bergin from TD Cowen.

I hear you on the overall complex demand environment. I wanted to dig in on the demand progression in the top accounts that you saw over the last three months, as well as just how you see the top clients, particularly the top five to ten, performing in the second half as you exit this year.

Despite the opticality of the clients for the top five, there is only one client which is declining, and this is basically a continuation of the trend which we saw before. Even this client decline is getting less, and we're getting to a more stable environment right now. But in general, it's exactly like we commented before. It is pretty stable.

Yes. And that top five client, this is the one that we talked about in the past, which is a European business information and media client.

Okay. All right. That's helpful. And then just on the GenAI front, can you dig in a little bit more on the progression of GenAI-related work as far as any rough quantification on the size of some of these programs and the mix of really the POCs that are moving into production?

It is pretty challenging, especially when we're trying to understand how some of our talents quantify this. It's very much all around the well. So that's why for our internal understanding and integration, we have kind of fewer GenAI-related projects and some influence revenue and so on. From this pure type of stuff, a lot of small POCs are now approaching high hundreds of thousands or low millions of dollars. This is already starting to happen. This is dozens in our case. It's very difficult to compare apples to apples when we're hitting some numbers for competition. At the same time, from an influence point of view, we started to go already to tens and upwards of hundreds of millions as well, very different even from six to nine months ago.

Okay. Understood.

Still, if you think about it, it's a very small portion of our revenue. The general trend is that when during the POC, there is a confirmation of potential ROI and excitement, then it comes back to the technical debt, which we were talking about during the last quarter. Companies realize that to actually get the benefit from this, it requires really significant investments and to go to some data modernization program which is much needed and requires much more depth. That's actually one of the showstoppers to real progression, because they are not ready yet.

Your next question comes from the line of Darrin Peller from Wolfe Research.

Could we discuss the sequential numbers briefly? The revenue guidance appears to be flat from the second quarter through the end of the year. I understand you're not incorporating any potential upside or changes. Additionally, there is typically some seasonality in the fourth quarter, though budget allocations have been challenging recently. Could you elaborate on your quarterly expectations? Also, please revisit the overall demand landscape. Where are you seeing client spending, especially in relation to the current macro environment? It seems that various IT services companies are focusing on addressing immediate needs. Can you provide insights on today's demand, particularly if the macro conditions remain slower for an extended period? What do you believe is resonating most with clients?

Yes. I'll just start with the more technical, I guess and I'll leave Ark to answer maybe the harder questions. From a Q2 to Q3, you'd have higher vacation and more build days. You should see a modest improvement but you should see some improvement in revenues based on what you can call kind of technical or seasonal factors. In Q4, it will depend on what type of vacation levels we see. Usually, you would see even slightly higher levels of vacation in Q4, lower build days. And then, of course, the question is going to be what type of furlough activity we see. Generally, there is a somewhat significant impact just due to seasonality. That's kind of what we're modeling at this time, again, is that generally, a very modest improvement from Q2 to Q3 based on seasonal factors. And then some degree decline unless, as you said, we see some type of budget flush or again, we're able to influence the level of vacations that employees take. Ark, do you want to talk a little bit about overall demand or where we're seeing?

I think our equipment is very much in line with the last quarter. As we mentioned last time, we don't think we can project the market in the current situation. It is very much similar. If a quarter ago, our projection range was much broader than today, we were saying that our expectation of good news was not confirmed. Our expectation for the great news actually didn't happen as well. We are narrowing, and it's a reflection of the type of projects in play right now. There is no big modernization talk. There are conversations about it, but it's not turning into reality. There is a lot of noise around GenAI which is not converting to big revenue as well. We're keeping the status quo on production systems, that's what we're focusing on improving, and looking for kind of one-off modernization plays where we can really bring value. It's very competitive and not necessarily decided for the client right now.

I would just add that we're working to change the trajectory in Europe, and we are beginning to see some better conversations and opportunities kind of appear there. Again, that's an area where we're looking to change the picture. The other thing I think you see in our fixed fee which continues to go up; we're continuing to explore and work with clients to have more of a committed model around what we'll deliver for a fixed fee or a fixed monthly fee, and that's a reflection of what we're trying to do to respond to customer needs and win more business.

Actually, one quick one just on hiring: do you anticipate needing to hire more if utilization stays in these ranges, or other types of efficiencies can help maintain?

There's certainly some programs where we're clearly working to include AI productivity improvements. But no, we would continue to hire and I think you'll continue to see hiring in the types of geographies we've been talking about which are more offshore certainly, with some in Latin America.

Your next question comes from the line of Jim Schneider from Goldman Sachs.

First of all, on the discretionary demand environment, it's not surprising to hear of the constraints given what the environment is out there. But what are your clients telling you about the conditions under which they would start to release more spending or be more aggressive with new projects in 2025? Is that tied to macro? Is that tied to more certainty around their AI strategy or other priorities they have internally in terms of IT spending?

We do believe that majority of the decision making is environment-related right now. As soon as the situation gets a little bit better, investment in general data infrastructure and cloud infrastructure which was delayed will be triggered, as everybody understands the impact of GenAI. Without fixing first this, it will be very difficult.

And then maybe just in terms of the margins, obviously, you delivered good growth and operating margin leverage in the quarter. Was that mostly driven by the mix of headcount shifting to India? Or are there other factors there besides the SG&A line? And then, I guess, going into '25, as we exit this year, what kind of further gross margin leverage do you expect to deliver or is this sustainable from here?

Yes. We're continuing to work on utilization. The improvement in Q2 was probably a combination of efficiency with SG&A and continued focus on improving utilization. What we've talked about over the last couple of quarters is that we continue to have an opportunity because we've got a fairly heavy pyramid still including in India. What we need to do is make sure that we're introducing more juniors into the mix, which generally has a broader sort of pyramid that improves profitability, overall also allows you to be a little sharper with pricing. The Q2 improvement in profitability was not driven by a shift in India. Again, it was more operational efficiency factors we're continuing to work on throughout the remainder of the year.

And in terms of the forward improvement there?

Forward improvement, again, is the work that we're doing on utilization improvement, reducing the bench, and ongoing efficiency in SG&A. So again, it's just a focus on certain areas of our operations that we think we can see some further reduction in spend, certainly as a percentage of revenue.

Your next question comes from the line of Jonathan Lee from Guggenheim Securities.

I want to get a better sense of how India is progressing. Can you help unpack the type of volumes you're seeing there and whether expanded presence has had any sort of influence on new types of demand or types of contract structures being utilized, especially as you think about the revenue and margin dynamics that are contemplating the outlook?

Yes. From the last time we spoke with you, Jonathan, is that India is likely to make a somewhat greater percentage of headcount by the end of the year. So last time, Ark and I were talking about something approaching 20%. We now think that India will be slightly above 20% by the end of the year. What you are seeing is a modest pressure on average bill rates as a result, and that probably is also shaping how we look at the second half. So it's not super significant, but I think last time we would say 19.5%, and right now we think that India is going to account for just over 20% of our headcount by the end of the year. We're continuing to see a modest gradual shift there, while at the same time, we are seeing improved utilization in our other areas of operation in Europe and Western and Central Asia. It's not as if all the demand is shifting to India, but we are seeing ongoing kind of increment in India.

Regarding the type of work, we consider that in India, there is pricing pressure. This is definitely a very objective component. At the same time, the type of work which we do is not changing much from location to location. EPAM has a reputation for more complex quality engineering solutions. We're also building a very strong data renewal occupancy. We're bringing everything that we do around GenAI and productivity improvement for SDLC. We're building a digital engagement practice. So it's very much in line with the broad EPAM, and the type of work is again very, very similar.

Thanks for the detail there. Can you unpack your comments on the lack of improvement contemplated in the outlook? I want to understand what that means for deals that have been signed but perhaps not yet launched or ramped. How much go-get or pipeline conversion is still required to achieve your outlook at both the high end and the low end?

Yes. And the last time we guided, we did talk about still expecting a very modest improvement in demand. What we're saying now is we don't see that improvement in demand. We try to be quite prudent with our guide. Clearly within this quarter and having set the full-year guide which is how we're thinking about Q4, it does encompass even things like potential reductions in demand due to cost reduction efforts at clients or that type of thing. We feel pretty confident that we have a little bit of downside as clients continue to be cost-sensitive. The upside probably would be in the lighter furloughs or maybe just a little bit of kind of budget openness in the remainder of the year. Our ability to also influence the level of vacation taken by employees will give us a little bit more capacity in Q4.

Your next question comes from the line of Ramsey El-Assal from Barclays.

It looks like your percentage of fixed-price contracts has been trending up and it's a little higher now than it's been at least going back quite a ways in our model. What is driving that mix shift away from time and materials work towards fixed price work? Is it geographic? Is it GenAI related? Are there any implications for margins when it comes to fixed price versus time and materials work?

Yes. You're correct that it has been trending up and probably will continue to trend up somewhat. It reflects the fact that we're beginning to try to address clients' needs in a way that's a little bit more traditional prepay. We have projects which are difficult to estimate. We clearly have that type of work but we are trying to sit with our clients and say we can deliver a certain program for a fixed amount of money or fixed amount of money on a monthly basis. The other opportunity with GenAI is to introduce productivity improvement and commit to a series of savings over a period. It can be net positive to margins. If we've misestimated or delivered poorly, it could be negative. Generally, with fixed fee, you do have the opportunity to improve profitability relative to time and materials because it just gives you more flexibility in how you deliver.

Okay. A follow-up for me on M&A. Given the buybacks in the quarter and the additional share repurchase authorization, is larger scale M&A off the table? Assuming you're still in the market for tuck-ins to plug capability gaps, what types of assets might you be looking to bring into EPAM?

Nothing is off the table. As always in the past, we constantly have conversations and opportunities for different sizes of acquisitions. The share buyback is actually very much a function of if it's going to happen or not. It's not a must condition; it's a direction we will be executing only if we think that we will ramp up with any other aspects of the business. If M&As are going to happen, we will adjust the numbers if necessary.

We'll do both. Our bias would be towards acquisitions. As Ark said, doing something somewhat larger is certainly not off the table.

Your next question comes from the line of Surinder Thind of Jefferies.

Just a question around the global delivery footprint. As you look ahead, if revenues were to remain stable, at what point do you think you'll get to your target delivery footprint?

When you say it will be stable, what do you mean?

I believe there was mention during the call that on-site utilization fell slightly short of expectations. This indicates a continued transition towards utilizing resources in lower-cost areas. There was also previous discussion about reduced demand in nearshore or Western Europe, leading to the natural reallocation of some of those resources to different regions. That was the point I was trying to make.

We do definitely move into global diversification from indications of stability and 24/7 on the growing global climate. We will be much more diversified than in the past. Right now, as we mentioned, we'd probably be the most balanced global direct company. That's a direction. What exactly the proportion of this is, it's much more difficult to answer because it would be a function of general demand. When you say, for example, that assurance in Europe is not so much in demand, it's in many ways, subject to the type of work and number two, the cost pressure. As soon as the market starts to come back to fix the technical debt, the modernization cloud and data program will accelerate again. The demand will come back for practically any region. The complexity and creativity of these engagements will become much more important, and pricing components will become less significant. It would influence the structure as well.

That's helpful. Then, when you think about all of the new talent that you're hiring, how do you differentiate or attract that talent in the sense that others have large delivery operations out of India? They have well-established connections to local universities, whereas I would argue you're newer to that region. I realize you've been there since 2015 but just on a relative basis.

If we're talking about India specifically, I think a couple of factors need to be acknowledged. We have an image of a different type of services company, much more quality engineering, and much more aligned with what people think about when they think of software tech companies. From a talent perspective, we compete with these types of companies, the same as we do with some captives that are trying to build high-end tech purchases in India. The image is already there. At the same time, we are also kind of an underdog in India, meaning that we have the opportunity to play differently in specific parts of the market, including bringing our training capabilities and different types of work.

Your next question comes from the line of David Grossman.

I have a couple of quick follow-up questions. If I remember correctly, you mentioned that the headwind from India and the mix shift in India would be around 200 basis points this year. Is that still a reasonable estimate? Also, do you have any initial thoughts on the potential magnitude of the headwinds for 2025?

Yes. I would say that 2% is generally correct. It probably has gone up slightly from when we guided at the end of the Q1 call. For next year, my guess is the headwind from India might be greater than 2%. That's how I'd respond to that.

Got it. And similarly, I know you've talked quite a bit about the lost clients. I think one was M&A and one was something else. I'm just trying to remember whether you quantified that headwind this year and next and when we come out against that headwind?

Yes. The one that is kind of the M&A exit is the one I usually refer to, which we called out at the end of the year. It was double-digit revenue, over $10 million a quarter, a significant number. The other one has been reducing their demand for our services, a gradual decline over time, as they face challenges around our Ukrainian footprint, coupled with some work they're bringing in-house. We still have demand from them, but there has been a gradual decline over time, and I think you'll continue to see that for the next couple of quarters.

Got it. One last thing, just on the DSO, Jason, I know it was up last quarter and it was up again sequentially. Should we see that or do that as maybe a macro dynamic affecting all your accounts? Or is this another way of providing better terms to remain more competitive? Or is there something else going on?

I would say that the 76 days was definitely a result of the last couple of days being on a weekend and we saw a significant amount of cash coming on the Monday and Tuesday, but that was here in Q3. However, as we've moved towards more fixed fee, that is impacting the DSO, because it does affect invoicing. You may see something closer to 74 days for the second half of the year. It's due to the shifting in the type of contracting we've been doing.

Operator, we have time for one more question.

The final question comes from the line of Jamie Friedman from Susquehanna.

Regarding your previous response, Jason, I found your insights on the fixed price to DSO quite intriguing. I'm curious if the growth in fixed pricing is connected to generative AI or outcomes-based pricing. That's my first question. Additionally, Ark, could you provide an update on what's happening in the life sciences sector? Jason mentioned it includes both healthcare and life sciences in his comments. It was a significant topic at the Analyst Day a couple of years ago, and it seems to be making progress now.

Fixed fees certainly reflect some experimentation with fixed fees with productivity that's GenAI driven. I don't think that's showing up in the numbers right now, but it may continue to show up in increased fixed fees, as clients are looking for us to step in and say we can deliver a certain program for a fixed amount of money or fixed amount on a monthly basis. We are also doing more business in the Middle East and that market tends to be more fixed fee oriented as well.

Regarding growth in life sciences & healthcare, we have been building more industry expertise in this area. From our conversation several years back, it was positive growth, while a couple of things when client situations changed. Right now, it's pretty positive. The concentration of data programs in our life sciences & healthcare business is quite high, which remains in demand today. Again, a combination of industry expertise which we've invested in, some level of consultancy, and again, data program proportion makes this beneficial for us. It's something we look at in other industries, how we can change the trend similar to what is happening here. I think it's time is over. As usual, thank you for joining us today. We believe that while the situation is as it is right now, we do believe that EPAM is focused on cloud data engineering and with GenAI, we are pretty well positioned for future growth when markets come back. We know we repeat this each time, but we believe it. We'll be ready for a comeback, and we will see how many quarters we will still have to wait for this. Again, thank you and let's talk in three months.

That does conclude our conference for today. Thank you for participating. You may now all disconnect.