KANZHUN LIMITED

Ladies and gentlemen, thank you for standing by and welcome to the Calsium Limited Second Quarter 2025 Financial Results Conference Call. At this time, all participants are in a listen-only mode. After the speaker's presentation, there will be a question and answer session. Today's conference is being recorded. At this time, I would like to turn the conference over to Ms. Wenbei Wang, Head of Investor Relations. Please go ahead, ma'am.

Thank you, operator. Good evening and good morning, everyone. Welcome to our second quarter 2025 earnings conference call. Joining me today are our founder, chairman and CFO, Mr. Jonathan Peng Zhao, and our director and CFO, Mr. Phil Yu Zhang. Before we start, we would like to remind you that today's discussion may contain forward-looking statements, which are based on management's current expectations and observations that involve unknown and unknown risks uncertainties, and other factors not under the company's control, which may cause actual results, performance, or achievements of the company to be materially different. The company caution you not to place undue reliance on forward-looking statements and do not undertake any obligation to update this forward-looking information, except as required by law. During today's call, management will also discuss certain non-GAAP financial matters for comparison purposes only. For definition of non-GAAP financial measures and the reconciliation of GAAP to non-GAAP financial results, please see the earnings receipts issued earlier today. In addition, a webcast replay of this conference call will be available on our website at ir.jp.com. With that, I will now turn the call to Jonathan, our founder, chairman, and CEO.

Hello, everyone. Welcome to our company's 2025 Secondary Funding Regulation Conference. Hello, everyone.

Thank you for joining our company's second quarter 2025 earnings conference call. On behalf of the company's employees, management team, and board of directors, I would like to send our sincere gratitude to our users, investors, and friends who have continuously believed in and supported us.

Today I would like to report on four matters.

First, our quarterly performance was good. Second, the supply-demand dynamics on our platform continue to improve. Third, we are making ongoing progress in AI. And fourth, our recent Hong Kong share offering and future shareholder return arrangements.

First of all, I would like to say this. According to the report, the company's net income of 21 million yuan has increased by 99.7%. Let me start with an overview of our financial performance.

In the second quarter, our company achieved a total revenue of RMB 2.1 billion, up 9.7% year-on-year. Our net income reached RMB 710 million, reflecting a 17.4% year-on-year growth, achieving a net profit margin exceeding 33%.

Excluding share-based compensation expenses and other incomes, such as investment banks,

Our adjusted operating profit was RMB 880 million, up 33% year-on-year. Share-based compensation expenses for this quarter decreased by nearly 10% quarter-on-quarter for the second consecutive quarter, amounting to RMB 230 million, with the ratio to revenue narrowed by about 5 percentage points year-on-year.

The scale effect brought by the business change and high-quality business model supports our high-quality growth of income and profit at the same time.

The operating leverage from economics of scale and the efficient business model supported our high-quality growth, characterized by simultaneous improvement in both revenue and profit.

From January 1 to July, the company has accumulated more than 30 million new users. From January to July, we cumulatively added over 30 million verified new users.

In the second quarter, the average verified monthly active users on the BOSS checking app reached 63.56 million, up 16.5% year-on-year. Consistent with user growth and penetration trends, brand-new contributions from blue-collar, lower-tier cities, and small- and medium-sized enterprises further increased compared to last year.

Second, during the second quarter and graduation season, our brand's recruitment and recruitment relationship has significantly improved. Specifically, at the recruitment end, the increase in the number of recruiters has significantly reduced the demand for recruitment. From June to July, the number of new student users has decreased by more than 20%. At the recruitment end, the recruitment demand for recruiters has increased,

During the second quarter and graduation season, we saw notable improvements in our platform's job supply and demand dynamics. On the job seeker side, incremental job seeking demand from fresh graduates moderated, with number of newly added verified graduates declining over 20% year-on-year in June and July. On the employer side, recruitment demand for fresh graduates also increased. The number of new job postings for fresh graduates grew by over 18% year-on-year for the same period.

This is also consistent with the trend of the recovery of the overall recruitment market demand. In July, the number of new jobs on the platform increased by about 20%. Whether it is the number of recruiters corresponding to the new jobs, or the number of recruits issued by recruiters is higher than in the same period last year. The improvement of supply and demand relations has also led to a clear decline in the number of new users.

This aligns with the overall trend of recovery in the recruitment market. In July, the number of newly posted jobs on our platform increased by approximately 20% during the year. Both the number of employers posting new jobs and the average number of jobs posted per recruiter were higher than the same period last year. The improvement in supply and demand relationships also led to a significant year-on-year decrease in the CB ratio for new users.

The improvement in supply and demand relationships also led to a significant year-on-year decrease in the CB ratio for new users.

Improved supply demand dynamics also drove positive changes in monetization.

The total paid enterprise customers in the 12 months ended June 30, reached 6.5 million, up 10% year-on-year. From industry perspective, blue-collar manufacturing experienced a short-term slowdown in April due to tariffs but resumed year-on-year growth from May onwards.

with growth rates continuing to outpace other industries. Urban service sectors also accelerated year-on-year growth in the second quarter. We observed a noticeable recovery in the internet industry, with the number of active job postings in the second quarter reaching a new high since 2021, led by product and technical growth.

Third, the company's continued progress in AI.

I will stick to the three perspectives, AI to C job seekers, AI to B enterprise users, and AI to management.

First, let's talk about AI to C. The AI interview training robot has made new progress. This robot is now taking on the role of recommendation. Under the premise that the applicant agrees, we used the data obtained during the interview to make recommendations for the applicant. The results of the participants' experiments have been improved.

First, AI to job seekers. The AI interview training robot has made some new progress. This robot now starts to play a role in recommendations. Upon job seekers' consent, we use data obtained during the interview process to recommend positions to job seekers, and users who participate in the experimental group have achieved higher efficiency.

AI supports user search to achieve new progress. For users participating in the test, AI can not only provide some explanations for the search results, but also provide more dynamic content, strategic planning for the job, and content such as simplification based on user requirements.

We continue to iterate AI-assisted user search. For users participating in the test, not only the AI can give more information for the research results, it can also provide dynamic content summaries, job search strategy planning, and resume optimization guides based on user queries.

Users in the experimental group gain more mutual achievement.

In terms of protecting job-seeker safety, we have applied AI to identify risky users. For instance, we have trained AI to recognize more subtle aggressive language and expressions that violate platform rules. Another example is that the AI tools we developed have made preliminary grants in identifying fake content tampered by other AIs. This is obviously a long-term and a challenging task, but we firmly believe that more people are in need of such functions.

Let's talk about AI2B. We asked many new founders

Ladies and gentlemen, please continue to stand by. The conference will resume shortly. Thank you for your patience. Can you hear me now?

Yes, we can hear you now. Please resume.

Okay, we start over. Where do we start? Let's talk about AI2B again.

It just stopped. Sorry. Let's talk about AI2B again. We provide AI support for the recruiters of many new startup companies and many new HRs. Next, about AI to enterprises.

We provide AI-assisted job posting optimization features for bosses of many newly established startups and junior HR. AI assists in posting tens of thousands of job positions on a daily basis. The key point here is how to prevent turning assistance into replacement.

It is difficult, but we must persist in doing so. We have extensively integrated AI to conduct experiments.

For instance, we use AI to enhance the understanding of recruiters' intentions, thereby helping them to select value-added services which are more suited to their needs. Recruiters in the experimental group have made more proactive purchases, and because they have chosen more suitable products, the repeat purchases have also increased.

to modify the development tools and the development process. A technical department of Beijing headquarters is currently producing 30% of the code by AI. A research and development base in another city is currently producing 70% of the code by AI. The result is that the speed of product development on the line has improved a lot. Therefore, within three hours, Now AI to management.

We promote the use of AI in research and development, transforming R&D tools and the processes. In a certain technical department at the Beijing headquarters, 30% of the coding is now AI generated. In another city, a newly established R&D department, 70% of the code is AI generated. As a result, The speed of product R&D iteration and launch has significantly increased, allowing us to explore more possibilities within the same time frame.

AI is increasingly helpful for customer service. AI is more and more helpful for customer service. AI is more and more helpful for customer service. AI is playing an increasingly important role in customer service.

It has achieved results in training new customer service staff, automatically inspecting customer service quality, and providing suggestions to recognizing and responding to customer emotions. This is crucial for improving user satisfaction and enhancing the well-being of customer service employees.

最后报告一下公司近期在香港增发和股东回报的安排。 公司在7月4日完成了22亿港币的香港公开增发。 主要目的就是为了提升公司在香港的流动性, The last one, we would like to report our recent Hong Kong offering and shareholder return arrangements.

The company completed a Hong Kong secondary share offer of HK$2.2 billion on July 4th. The primary purpose was to enhance the liquidity in Hong Kong line, allowing more investors in the Hong Kong stock market to understand and participate in the company's trading. The offer has achieved positive results with a significant increase in Hong Kong stock trading volume compared to the pre-offering levels.

In terms of shareholder return, the board approved two shareholder return plans today. The first is to announce an annual dividend.

The company plans to implement a dividend every year.

This year's dividend amount is $8,200. The second is to launch a new phase, which is also our fifth share return plan. From August 29th, within the next 12 months, the company will buy back more than $2.5 billion in shares. Regarding shareholder returns, the Board of Directors approved two shareholder return proposals today.

First, an annual dividend policy was adopted. The company plans to pay out annual dividends going forward with dividends of US$80 million for the current fiscal year. Second, the Board of Directors approved two shareholder return proposals today. A new share repurchase program and a fifth cumulative perspective is launched. The company intends to repurchase up to $250 million of its shares over the next 12 months, starting August 29. We believe this fully demonstrates the company's sincerity in actively rewarding shareholders and sharing the benefits of our sustained growth with all investors.

That concludes my part of the call.

I'll now turn it over to our CFO Phil for the review of our financials. Thank you.

Thanks Jonathan. Hello everyone. Now let me walk through the details of our financial results of the second quarter of 2025. we continue to achieve high quality results in this quarter represented by solid revenue growth and further improve the profitability the revenue growth this quarter was primarily attributed to the continued expansion of our user base with the number of paying enterprise customers in to 6.5 million over the trading 12 months ended June 30th. As the recruitment market demand has gradually recovered since the beginning of this year and the job seeker recruiter ecosystem has improved, the willingness of enterprise clients to pay has been rising. Among them, the recovery in recruitment demand from small and medium-sized enterprises has been more pronounced, driving a quarter-on-quarter increase in the revenue contribution from SMEs. ARPU maintained stable and modest growth, mainly benefiting from the expansion of paying amount from key accounts. Moving to the cost side, Total operating costs and expenses decreased by 7% year-on-year to RMB $1.5 billion this quarter. Share-based compensation expenses dropped by 24% year-on-year and 9% quarter-on-quarter to RMB $230 million, shrinking for the fourth consecutive quarter. Excluding share-based compensation expenses, adjusted income from operations grew by 33% to RMB $881 million, and our adjusted operating margin in the quarter reached 41.9%, up by 7.5 percentage points year on year, hit a record high. Cost of revenues. decreased by 3% year on year to RMB 307 million in this quarter, mainly due to the decrease in operational employee related expenses as a result of improved operational efficiency as we continue to engage AI in our daily operations. Gross margin went up by 1.9 percentage points year on year to 85.4% Sales and marketing expenses decreased by 23% year-on-year to RMB 420 million during this quarter, primarily driven by decreases in advertising and marketing expenses and employee-related expenses. However, our strong brand recognition enhanced marketing efficiency and superior user engagement guaranteed that we can still maintain robust user growth momentum. Our R&D expenses decreased by 6% year-on-year to R&D 416 million in this quarter. This decrease was primarily driven by reduced public cloud service fees related to AI. Our G&A expenses increased by 19% to RMB 311 million in this quarter, primarily due to an increase in employee-related expenses and investment in new initiatives. Our net income increased by 70% to RMB 711 million in this quarter. with adjusted net income increased by 31% to RMB 941 million. And the margins also expanded significantly and reached record highs. Our net margin improved by 12.1 percentage points year on year to 33.8%, while our adjusted net margin reached 44.8%, up 7.3 percentage points year on year. Both of these two margins have maintained sustainable improvement over the past three consecutive quarters. Net cash provided by operating activities reached RMB $1,052 million in this quarter. up 21% year on year. As June 30, 2025, we continue to maintain a strong cash position of RMB 16.0 billion. In July, we completed the share offering of 34.5 million Class A ordinary shares at HK$66 per share, comprising a Hong Kong public tranche and an international tranche. Net proceeds from this share offering amounted to approximately HK$2.2 billion. This offering, on one side, improved our Hong Kong line liquidity and broadened our shareholder base. On the other side, further strengthened our cash position, gave us both the strategic flexibility and financial capacity to pursue long-term growth initiatives and enhance our shareholder returns. One new initiative Jonathan just mentioned is that our board of directors has just approved the adoption of an annual dividend policy with a dividend amount of 80 million US dollars for the fiscal year of 2025 combined with a renewed 250 million US dollars share repurchase program our commitment to shareholder returns continued to enhance and for our business outlook just like we communicated before we expect our revenue growth to re-accelerate starting this quarter along with the recovery of recruitment market momentum for the third quarter of 2025 we expect our total revenues to be between RMB 2.13 billion and the RMB 2.16 billion. A year-on-year increase of 11.4% to 13%. This concludes our prepared remarks. And now we would like to answer questions. Operator, please go ahead.

thank you as a reminder to ask a question please press star one and one on your telephone and wait for your name to be announced to withdraw your question please press star one and one again once again please press star one and one on your telephone and wait for your name to be announced to withdraw your question please press star one and one again thank you we are now going to proceed with our first question And the questions come from the line of Eddie Wang from Morgan Stanley.

Please ask your question. the food delivery war that started in the second quarter led to the increase in the recruitment needs of the service industry. I also want to ask Mr. Zhao if the recovery of recruitment needs can continue. How do we see the development of recruitment needs in the third quarter? This is the first question. The second question I want to ask is that Mr. Zhao just mentioned that we have a considerable proportion of codes in the R&D department that may already be generated by AI. I want to ask, how does our company see the future of AI for the entire Biling recruitment, especially the need for programmer recruitment? I will translate these two questions myself. Thank you, management, for taking my question. I have two questions. The first one is the recruitment demand, the recovery we witnessed on BOSS platform in the second quarter and July. Is there any different driver for such a recovery? this time versus before? For example, has the food delivery battle led to the surge in service industry blue collar recruitment amount? Do you think the recruitment amount in second quarter is sustainable or not? How's your view and outlook for the third quarter? And the second question is that you mentioned previously that your R&D department, most of the coding has been generated by AI. So what's your view on the AI's impact on the white-collar recruitment, especially on hiring demand of the programmers?

Thank you.

The recovery of the recruitment demand this time compared to the previous one has a small and a big feature. The small feature is The position of the Internet industry, as I just mentioned, has reached a new height since 2021. One of the major features is that small and medium-sized enterprises are recovering faster.

Thank you for your question. About the recovery trend of the recruitment market, compared to before, we saw small features and big features. The small one is the job postings from the internet sector have recovered to a new high since 2021, as we just communicated. And the big one is that the smaller size or micro-sized enterprises have been recovering much faster.

One data to share with you is that the company with less than 20 employees

In the second quarter, its revenue contribution goes up to almost 20%, which is representing the highest growth rate among all different types of companies.

Let's expand the scale a little bit and look at companies with less than 100 employees. The new number of jobs, in comparison, is also the average level of jobs that exceed the platform.

Let's expand the size a little bit to companies with less than 100 employees.

The year-on-year growth rate of new job postings by those companies have also significantly exceeded the platform's overall level. The impact I see is something that can be ignored. Although the media is more concerned about the foreign trade war, from the perspective of the jobs that can be truly contributed, the position of the hands of the foreign traders is extremely low in the supply of our entire position.

about your question of the impact of the food delivery competition our observations is impact is quite minor or natural so because the data shows that the job postings related to riders or food delivery guys has been quite small among all of our job postings. And we also haven't observed any higher than average level of revenue growth from those jobs.

关于您关心的就是这次的恢复, 招聘需求的恢复是否可持续, 我们的看法是乐观的。 我有以下依据。 首先就是说, 我们在以前的Earnings Conference Call上曾经报告过好几次, 从去年的7月, 招聘者跟求职者的比例是比较差的。 去年的11月, 这个比例已经恢复到比前年的11月好。 About your third question, whether this recovery could be sustainable, our view is positive.

I have several evidence for you. First one is we have communicated before that the very poor job seeker to recruiter ratio starting last July, the situation has been improved since last November, November 24, which by that time has recovered to the same period of November 23. After that, the job seeker to recruiter ratio or dynamics have continued to improving.

And in the second quarter, we have seen a much more obvious recovery for that number, which we have just communicated.

From my perspective, from daily operation, our business growth rate in the third quarter is expected to further accelerate compared to the second quarter, which I have quite confidence with. And that's our answer to the question. And, Operator, let's move on to the next question.

Sure. We are now going to proceed with our next question. And the questions come from the line of Wei Qiang from EBS. Please ask your question.

Thank you. On the basis of the current high-tech number, how should we consider the profit and loss trend next year and in the future? Based on our stable profit and loss and cash flow, what are our most important investment directions in the next few years? Thank you, management, for taking my questions. I have two questions. First is our margins continue to expand this year to a very high level. So given such a high base, how should we think about the margin trend in the next year and beyond? And also considering our healthy and stable margins and cash flow, what do we consider as the most important investment areas going forward? Second, it looks like some startup companies are ramping up advertising investment recently. Does it affect our marketing and user acquisition costs? How do we assess the impact on the competition landscape in blue collar segment and online recruit market overall, as well as our competitive modes. Thank you.

So thanks for the question. And I'll answer the margin questions first. You are right that our margin continues to improve. We think that this is mainly related to our business model. Because we run an online recruitment marketplace, the scale effect brought by our company's business model is significant. We believe this is the fundamental reason for the continuous improvement of our profit margins. Under the company, in the past several quarters, we implemented effective cost control to make sure we have know we put our focus on the high quality part of growth which kept our growth you know kept our cost growth rate lower than that of a revenue so with that our you know along with our revenue steady growth the direction of a gradual improvement of our profit margin is quite clear, and this is kind of definite. But we believe the margin improving is a long-term thing, and it should run step by step, not all of a sudden, not grow too fast or too high in short term. So basically, in terms of the areas that we would like to invest, we will continue to invest into our business. And our future investment priorities remain consistent with our previous ones, mainly focusing on R&D, innovation, and new business initiatives, et cetera. And you also can see that we generated a very healthy cash flow. So in this quarter, we generated more than 1 billion RMB operating cash flow. Actually, this is the conservative two quarters. So the first quarter, the operating cash flow is also about 1 billion. So we already had two conservative quarters with over 1 billion operating cash flow. And with such healthy cash reserves, we will mainly use our cash at talent development, probably like overseas expansion, and more importantly, shareholder return programs in the future. So that's my comment related to the margin and our use, how to use of cash in future.

Jonathan can answer the second question.

We've seen many of our colleagues advertise in specific seasons, including some mature companies and startup companies.

So far, I think our impact is relatively small.

About the second question, yes, we do observe that many of our peers will do some advertising marketing events in certain cities, including both mature companies or startups. But so far to now, the impact to us is quite limited.

First of all, at this stage, the meaning of starting an advertising station is very, very small. The proportion of our own market promotion fee compared to revenue is decreasing. But in terms of absolute value, as far as I know, our market promotion fee in terms of absolute value is still the largest in the entire industry, but it is indeed lower than our income level. I think this is

But my observation or my understanding is that at this current stage, a marketing wall, the meanness of a marketing wall is very small. I want to clarify one thing that even though our marketing expenses as percentage of revenue is continued to declining, but on an absolute Among the spaces, we are still investing in the largest amount of marketing and advertising spaces among the industry.

At the same time, because we have established a strong dual network ecological effect, I am spending the same money. . . . . . .

At the meantime, due to our very powerful double-sided network effect, our very high user acquisition efficiency, so we can maintain a very robust user growth while our user acquisition costs still kept at a quite low level. And also, our user retention is the highest among our peers. So to sum up, we have very strong marketing investment. Our user acquisition efficiency is high. Our user retention is high. So that's why I said our current stage is quite low to start a marketing rule.

This is my answer to your second question. So, in fact, just now, Eddie Wang asked about the problem of AI-attached programmers. Let me talk about this question. What we see is that I believe that this is the same for many tech companies.

And we would like to answer the second question from Eddie about whether AI will replace programmers. So from our own situation is that yes, our requirement for entry level programmers have been slowing down, but we are still hiring. So we are focusing on those people who have more potentials, who are much more smarter, can break down the questions and looking from a bigger picture perspective.

So I think that's also happening in many other technology companies.

In terms of programmers, my change is that I actually spend more salary to hire a high-income actual level programmer. And I'm hungry to hire such a person, because such a person is always very competitive. So the manpower and physical strength I spend on recruitment will also increase.

I believe the situation I encounter will also be the situation of many technical companies.

And so what I'm facing to recruiting the entry-level programmers, I'm spending more money, more cost, recruitment cost, to hire the people with more talents or more potential. So we will, actually we might somehow increase the recruitment cost for those type of people. I think that's also happening to many other technology companies.

So this is how I think about it.

In the past, a high-level programmer had a bunch of people. High-level programmers did complex tasks, and people did simple tasks.

Now, a high-level programmer can have an AI.

This is a change.

So my thinking on this issue is that in the past, the normal structure is one senior programmer with some junior level team to do more simplified jobs. Now it will turn into one senior program of senior technical guys with AI, but the salary we need to pay and the hiring efforts we need to invest to hire that senior guy actually increases.

出一程序员的话,在整个百里,我们国家有一个多亿的百里,脑力劳动者。 And the junior engineer or junior programmer has a percentage of overall white-collar employees, which in our country might be more than 100 million.

my gut feeling is less than 2%. So that change to the impact of hiring of white collar is very limited.

Okay.

And that's all of our answer to the questions. And operator, let's move on to the next.

Sure. We are now going to proceed with our next question. And the questions come from the line of Timothy Zhao from Goldman Sachs. Your line is open. Please go ahead.

Hey, Mr. Zhao, good evening. Thank you for accepting my question. I also have two questions I would like to ask. The first one is that I would like to ask Mr. Zhao to introduce in detail some of the user scenarios on the AI side, on the recruiters' end, and on commercial products. We also see that the company has recently used AI to develop some small programs or apps with different functions. I would like to ask Mr. Zhao to share with us what we think is more potential and what our future commercialization plans are. The second question is about our recent financing in the Hong Kong stock market. I see that the company has also announced the resumption and resumption today. Thank you, Benjamin, for taking my question. Two questions from my side. First is regarding the AI application or AI features. if management can introduce more detailed about the specific use scenarios on the enterprise side and what kind of commercial product that we are thinking of. Also, we note that recently the company leveraged AI to develop mini programs or applications of different features. Just wondering from your perspective, which ones have the more commercial potential? And secondly, it's regarding your recent financing activities in Hong Kong markets. And I'm very glad to see you also announced the share purchase and annual dividend policy this year. Just wondering if management can share what is your thoughts on the capital markets going forward. For example, what are your plans to improve the liquidity in the Hong Kong markets? And what are the detailed dividend plans that you have in mind? Thank you.

AI has a great potential in our work.

I just made some introductions. In fact, we are conducting a wide range of new product development and experiments.

I can also give some examples.

Actually, yes, we do have some good progress on the AI product development side. Actually, we have been broadly to do some new product launches and development. So I can give you more examples.

Let's start with an agent, an intelligent body. We made an intelligent body for recruitment called Hammer. Now, we are using more than 500 recruiters every day. This is called Hammer Agent. To put it simply, recruiters can communicate with Hammer, and then complete most of the tasks that recruiters have to do on a daily basis, until they can complete a task with the recruiters. The first one is an agent.

We developed a recruitment agent called Hammer. Currently, we use it with approximately 500 recruiters daily. The recruiters can complete the majority of their work on our platform by simply interacting with Hammer through dialogue. Of course, this part should have the jobseeker's consent. And sometimes jobseekers might say no, but if the jobseeker is okay, the Hammer will complete its job until to achieve an exchange or resume or content formation or achievement on our platform.

We will use a very long period of time, patience, and investment to deal with Hammer. So far, I think, instead of using a new technology to verify what we already know about the Jopping Industry, it would be better to say that we have already felt that Hammer will take us to a higher position.

we will attach very great patients to iterate this agent hammer. So rather than saying that we are using new technology to validate what we have already known in the tree, hammer actually will lead us into even greater technology.

We are more restrained on the hammer, but another example is that our experiment will be more aggressive.

And

Our program on Hammer is actually quite restrained. But the second one, we have been more aggressive. So the second example is where the job seekers have more tolerance towards this product, and the recruiters, it has higher leverage in terms of the negotiation progress. So it's actually a product we designed for large state-owned

enterprises which have just launched. Currently, this system allows for customization of digital human avatar.

interview questions and interview reports. It also supports AI-powered follow-up questions, multi-modal candidate emotion recognition and image recognition. Using this technology, we have supported over 20 AI-powered interview events for recent graduates at over 10 large state-owned enterprises, attracting nearly 30,000 participants. And the response has been positive so far.

There are many cases like this in our company.

I believe that this is the case in many companies. I actually think that AI is a combination of the existing product technology business model and the internal management of the company. It is still in the early stages. I believe that the combination of new technology and old business, over a long period of time, is revolutionary technology.

This combination is a form of bourbon movement. The real killer application, the real winner, should be gradually generated in a large number of bourbon movements. We have the patience and commitment

I think that we have a lot of those examples, and that also should be quite common in many other companies. Actually, I felt that the combination of AI technology and the current products application in technology management and the daily operation is still at an early stage. So the new technology combined with all the business for this matter, the more revolutionized the product is, so that it will be more like a ground activities movement. So the killer level applications will be generated among a lot of those ground movements. So we have enough patience and we have enough assets to waiting for that kind of application to be born.

So at this stage, we care a lot about AI. And I think its fundamental principle is to treat it as a bad exercise.

So the AI everybody cares about, we are also quite care about. And the fundamental principle is just treating it like a brown movie. The next stage is AI will be everything in our daily life.

And about the money, I would like to ask Phil to answer the money question. Okay, so I'll answer the second question. So as you know that the company has a very healthy cash balance on hand. We have more than 16 billion RMB cash reserve. And in terms of the operating cash flow, every quarter we have incremental more than 1 billion RMB inflow. So basically the company does not need to raise money from the market. So our capital activity, our fundraising activity in July, the purpose of that activity is to improve our Hong Kong line liquidity. Because in the past, we got public listed in Hong Kong through the by introduction without issuing new shares. So that made us with a very poor liquidity in our Hong Kong stock trading. So we want to solve this problem. And we think that solving this problem could be both benefit to the company and benefit to our investors, to our shareholders. So that's why we took a very rare approach to, you know, launch a public offering in Hong Kong. So the offering was very successful and all the participating investors all made money and our Hong Kong line liquidity, you know, realize a breakthrough and our liquidity started to improve since the offering. So what I'm saying is that the company we consider shareholders as our partners and we consider the shareholder return very important topic and in the past we mainly use shareholder repurchase program to return cash to the shareholders. We totally already launched four phases and the total purchase shares total amount more than about 400 million US dollars. So this time the company renewed our share repurchase program and also we announced our annual dividend policy. We make this as a regular routine and make it, you know, an annual thing. So basically every year we will consider our operating situations and pay dividend to our shareholders. And this year for the first, you know, For the fiscal year of 2025, we announced $18 million U.S. dollar for the dividend for this year. And our Hong Kong line, you know, liquidity, you know, because of the share, the public share offering, we raised $2.2 billion Hong Kong dollar. That's roughly... 280 million US dollar but compared with our renewed share repurchase program which is 250 million plus our 80 million annual dividend so our announced amount is already higher than the amount that the way raised in our earlier public offering so that also showed that we had a good commitment to our shareholders' returns. And one last thing related to our shareholder returns is that our share-based compensation as an expenses, we received the comments from shareholders that say share-based compensation was once a little bit high, and we controlled these expenses and from the results you can see that in the past several quarters these expenses continued to decline in terms of total amount and in terms of percentage to the revenues. So all above are just, you know, evidence or, you know, some kind of things that show our attitude towards the shareholders' return. And we consider this as a very important thing, and we will continue to do that in the future.

Okay, so that's all of our answers to today's questions, operator.

Due to time constraints, this concludes today's question and answer session. At this time, I will turn back the call to Wenbei for any additional or closing remarks.

Thank you again for joining us today. If you have any further questions, please contact us directly. Thank you.

This concludes today's conference call. Thank you all for participating. You may now disconnect your lines. Thank you.