AI recruiting is moving from résumé automation toward end-to-end hiring agents, but enterprise buyers increasingly want evidence that the technology produces measurable results. Moka AI, formerly MokaHR, says its Moka Eva recruiting agent cut per-store time-to-hire from 15.4 days to 3.8 days at a major international coffee and tea chain operating more than 20,000 locations. The company says the deployment also reduced cost per application by 60% and manual recruiting intervention by more than 80%.
The pitch for AI in recruiting has changed.
A few years ago, vendors largely sold automation around résumé screening, candidate sourcing and interview scheduling. Now, enterprise HR teams are asking a harder question: What happens to hiring metrics after an AI agent takes over parts of the workflow?
Moka AI’s latest deployment figures are an attempt to answer it.
The company’s Moka Eva is designed as an AI recruiting agent that operates across multiple stages of hiring. According to Moka AI, it can source candidates, screen applications, contact applicants to confirm interest, conduct AI interviews and assessments, and deliver shortlisted candidates to hiring decision-makers.
At a large international coffee and tea chain with more than 20,000 locations, Moka AI says the system reduced per-store time-to-hire from 15.4 days to 3.8 days—a 75% reduction. The company also reports a 60% reduction in cost per application and more than an 80% reduction in manual intervention compared with the customer’s previous process.
Those numbers matter because distributed hiring is one of the more difficult environments for conventional recruiting systems. A retailer or restaurant chain can have thousands of locations, frequent frontline hiring and large variations in recruiter and manager behavior.
The technical challenge is not simply processing applications faster. It is maintaining a consistent assessment process as hiring volume scales.
From ATS automation to an AI recruiting agent
Moka Eva sits in a rapidly developing category of AI recruiting technology that attempts to automate workflows traditionally split among recruiters, coordinators and hiring managers.
The platform uses a unified rubric for screening and interviewing, according to Moka AI. The company says its assessments have been consistent with human reviewers more than 92% of the time across its deployments.
Moka AI says Eva has screened more than 5 million resumes and assisted with more than 600,000 interviews, supporting more than 10 languages.
That scale is significant, although the figures are company-reported and do not independently establish that AI decisions are superior to human hiring decisions.
The distinction is important for HR leaders. A system can produce consistent evaluations without necessarily producing better evaluations. Consistency depends on the quality of the hiring rubric, the relevance of the screening criteria and the extent to which human recruiters review exceptions.
Moka’s proposition is therefore less about eliminating human judgment than standardizing the work that happens before a hiring manager makes a final decision.
The economics are becoming the selling point
Moka AI’s deployment comes at a time when the economics of enterprise AI are under greater scrutiny.
Aon found that 74% of organizations across Asia-Pacific have either deployed or are piloting AI programs. Yet only 21% said they could effectively recruit and retain enough AI talent, highlighting the workforce constraints that can make automation attractive in the first place.
At the same time, enterprises are becoming more demanding about the return on AI investments.
KPMG’s Q2 2026 AI Pulse found that 53% of organizations surveyed were deploying AI agents, compared with 55% in the previous quarter. The research also found that organizations orchestrating multiple AI agents across workflows doubled from 9% to 18%, suggesting that enterprises are moving from isolated experiments toward more connected agentic systems.
That trend is relevant to recruiting. If an AI agent can move a candidate from application to screening to interview without requiring a recruiter to intervene at every stage, the value proposition becomes operational rather than experimental.
But the economic calculation also includes inference costs, integration, oversight and the potential cost of bad hiring decisions.
WhatsApp shows where the model gets interesting
Moka Eva is also designed for channel-native recruiting.
For a sportswear and sporting-goods retailer in Southeast Asia, Moka AI says the agent handles candidate screening and interview booking entirely through WhatsApp, saving the recruiting team 685 hours per month.
That use case points to an underappreciated issue in frontline hiring: candidate engagement.
For many hourly or distributed workers, completing a conventional application workflow can be inconvenient. A recruitment process conducted through a familiar messaging platform can reduce friction between application and interview.
It also illustrates why AI recruiting agents may eventually compete on workflow orchestration rather than simply model intelligence.
The value comes from connecting candidate communications, screening logic, scheduling and recruiter workflows into one process.
Asia-Pacific is becoming an important testing ground
The timing is particularly relevant in Asia-Pacific, where hiring organizations are simultaneously accelerating AI adoption and confronting workforce capability constraints.
Aon’s 2026 Human Capital Trends research found that APAC organizations are ahead of the global average in AI deployment or piloting, with 74% reporting activity in this area. But the study also found a gap in AI talent availability, with only 21% of APAC respondents saying they could recruit and retain sufficient AI talent.
The result is a market where AI recruiting tools can serve two purposes.
The first is productivity: reduce repetitive recruiter work and shorten hiring cycles.
The second is standardization: create a repeatable hiring process across countries, locations and recruiting teams.
That second objective may ultimately prove more strategically important.
A company with 20,000 locations does not have 20,000 different definitions of what constitutes a qualified candidate—or at least it should not. Yet decentralized hiring often produces precisely that variation.
An AI recruiting agent can enforce a common rubric at scale, provided the organization has designed the rubric appropriately.
Human oversight remains the critical variable
There is a natural temptation to interpret an end-to-end recruiting agent as a replacement for recruiters.
That would be too simplistic.
Recruiting decisions involve contextual judgment, accommodations, candidate questions, unusual career histories and employment-law considerations that cannot always be reduced to a standardized score.
KPMG’s research reinforces the importance of human oversight as agentic AI moves into operational workflows. The firm identifies human-oversight skills among the leading challenges to deploying AI agents.
For HR organizations, that suggests a different operating model: AI handles high-volume, repeatable tasks while recruiters focus on exceptions, relationship management, candidate experience and consequential decisions.
That model is increasingly visible across HRTech.
Platforms from Workday, SAP, Oracle, Microsoft and specialist recruiting vendors are incorporating AI into talent acquisition, while newer agentic systems are attempting to automate entire sequences of work.
Moka AI’s deployment data offers one example of where that category is heading.
The question for enterprise buyers will not be whether an AI recruiting agent can perform individual recruiting tasks. It is whether the agent can deliver measurable improvements in time-to-hire, cost, candidate conversion and recruiter productivity without sacrificing fairness, transparency or human accountability.
For a market increasingly skeptical of AI pilots that never reach production, those metrics may become the real competitive differentiator.
Market Landscape
The AI recruiting market is shifting from point automation to workflow orchestration.
Traditional applicant tracking systems remain the system of record for requisitions, candidates and hiring workflows. Newer AI recruiting platforms are attempting to sit on top of—or integrate with—those systems to automate sourcing, screening, interviewing and scheduling.
Moka AI’s results arrive as enterprises become more pragmatic about agentic AI. KPMG reports that AI-agent deployment remains above 50% among its surveyed organizations, while multi-agent orchestration is growing.
For HR leaders, the competitive landscape increasingly includes three categories:
- Enterprise HCM platforms, including Workday, SAP SuccessFactors and Oracle, which are embedding AI into broader HR suites.
- Recruiting specialists, which focus on sourcing, assessment, interview intelligence and candidate engagement.
- Agentic recruiting platforms, which aim to execute multiple hiring steps autonomously rather than simply assist recruiters.
The distinction between these categories will likely become less clear as established HCM vendors add more autonomous capabilities.
The strongest platforms will need to demonstrate not just automation rates, but measurable business outcomes and responsible deployment controls.
Top Insights
- Moka AI says Eva reduced time-to-hire 75% across a 20,000-location retailer, highlighting AI agents’ potential for distributed enterprise recruiting operations.
- The recruiting agent automates sourcing, screening, candidate calls, interviews and shortlisting, shifting AI recruiting from individual tasks toward end-to-end workflow orchestration.
- Moka AI reports 60% lower application costs and over 80% less manual intervention, giving HR leaders concrete metrics for evaluating recruiting automation.
- APAC’s 74% AI deployment or pilot rate shows strong adoption, while limited AI talent availability increases pressure to automate repetitive workforce processes.
- As AI agents scale, human oversight remains critical because standardized recruiting workflows must still account for fairness, exceptions, candidate experience and accountability.
Join thousands of HR leaders who rely on HRTechEdge for the latest in workforce technology, AI-driven HR solutions, and strategic insights
