Artificial Intelligence has become the headline feature of almost every HR technology platform. Yet in many cases, the conversation is still centred on roadmaps, prototypes, or future possibilities.

At Seamless Technologies, we wanted to test what AI could actually do for HR, not just talk about it. We’re a technology company building infrastructure for work and finance, designed to empower working Africans and the businesses that employ them, and SeamlessHR, one of our anchor products, is where this case study is centred.

Over the past few months, our engineering and product teams have deployed three AI-powered capabilities into production for a select group of SeamlessHR customers across recruitment, payroll, and employee self-service. Instead of chasing AI for its own sake, we focused on fixing the repetitive, time-consuming tasks that HR professionals deal with every day.

Here’s what we’ve built and why.

The Problems We Set Out to Solve

Across hundreds of conversations with HR and payroll teams, we consistently heard the same frustrations:

• Recruiters spend hours reviewing and manually entering information from CVs before they can even begin evaluating candidates.

• Payroll administrators frequently receive urgent questions from executives who need payroll figures immediately, yet finding the right answers often means exporting reports and searching through spreadsheets.

• Employees repeatedly contact HR with routine requests that could be completed through self-service if the experience were more intuitive.

These aren’t isolated challenges. They represent thousands of hours of administrative work every year, time that could be spent on more strategic initiatives.

Our goal wasn’t to integrate AI to replace HR professionals. It was to remove unnecessary manual work so they can focus on higher-value decisions.

AI for Recruitment: Faster Shortlisting, Better Hiring

Our Recruitment Management System has long enabled recruiters to upload CVs and automatically extract candidate information. With the introduction of AI, we’ve taken that capability several steps further.

Traditional document parsing works well with structured CVs, but recruitment teams often deal with a wide variety of layouts, writing styles, formats, and inconsistencies. 

• AI allows our CV parser to better understand context, interpret unstructured information, and accurately identify key details such as work experience, education, skills, certifications, and career progression, even when candidates present them in very different ways.

The result is higher extraction accuracy, fewer manual corrections, and greater confidence in the candidate data that recruiters rely on throughout the hiring process.

• We’ve also enhanced candidate ranking with AI. Once a hiring manager defines the criteria for a role, the platform goes beyond simple keyword matching to evaluate applicants more intelligently against the required qualifications, experience, and skills. 

Recruiters receive a prioritised shortlist that helps them identify the strongest candidates faster while still retaining full control over hiring decisions.

Combining more accurate CV understanding with AI-assisted candidate evaluation means recruitment teams spend less time validating data and screening applications, freeing them to focus on engaging candidates, conducting interviews, and making better hiring decisions.

AI for Payroll: Insights on Payroll When You Need Them

Payroll teams don’t always need another report. Sometimes, all they simply need is a concise answer.

To enable payroll teams to become more productive, we built SeamlessIntelligence, an AI-powered payroll assistant available directly within the payroll approval workflow on the payroll management system.

Instead of navigating multiple reports, payroll teams can ask natural-language questions such as:

• “What’s our total payroll cost this month?”
• “How many employees were paid this cycle?”
• “What’s the total overtime payout?”

The assistant retrieves the relevant information and returns clear, conversational answers in seconds.

Payroll data demands a high degree of accuracy, so we built SeamlessIntelligence on Claude by Anthropic.

Adopting the Claude by Anthropic model gives SeamlessIntelligence the ability to reason carefully over payroll data while presenting results in plain language that finance, payroll and HR leaders can immediately understand.

AI for Employee Self-Service

Employees shouldn’t need to understand HR processes before using the self-service platform available to them. They should simply be able to ask for what they need without needing to contact HR.

That’s the thinking behind Samira, our AI-powered Employee Self-Service assistant.

Samira combines conversational Artificial Intelligence with secure access to employee data, enabling employees to complete common HR tasks using natural language instead of navigating menus or searching knowledge bases.

Whether an employee needs information about leave, payroll, personal records, or other HR processes, Samira understands the request, retrieves the relevant information securely, and guides the employee through the appropriate action.

Behind the scenes, Samira leverages Anthropic’s Claude models together with the Model Context Protocol (MCP), allowing secure, permission-based access to live organisational data without requiring bespoke integrations for every individual use case.

Building Responsibly Before Scaling

Launching AI features into production isn’t simply about making them available. It’s about ensuring they deliver real value. Before opening up our AI features to all customers, we invited selected customers to participate in a structured pilot programme.

During the pilot programme:

• Recruitment teams created jobs with customised ranking criteria and evaluated how accurately AI ranked applicants.
• Payroll administrators tested SeamlessIntelligence using real operational questions, assessing both the speed and quality of responses.

Their feedback has helped us refine the experience before broader rollout. We believe customer validation should come before marketing claims.

As the pilot concludes, we’ll publish measurable outcomes including recruiter time saved, AI response accuracy, customer adoption and user feedback.

Our Philosophy on AI

The general conversation around AI often focuses on what’s technically possible. However, at Seamless Technologies, we prefer to focus on what’s genuinely useful.

For us, AI isn’t about replacing HR professionals or automating every interaction.

It’s about eliminating repetitive work, simplifying access to information, and enabling people to spend more time on decisions that require human judgement.

Every capability we build follows the same principles:

• Solve a real business problem.
• Keep humans in control.
• Protect customer data through strong security and governance.
• Validate with customers before scaling.

What’s Next?

The capabilities spotlighted in this case study represent only the beginning of what we’re building with Artificial Intelligence.

Our product and engineering teams are already developing an AI-native reporting and analytics experience that will allow business leaders to explore workforce insights using natural language across their HR and payroll data.

As with every feature we’ve released so far, we’ll validate it with customers before making it broadly available. We believe the future of HR technology isn’t about adding AI everywhere.

It’s about applying AI thoughtfully where it delivers measurable business value.

And that’s exactly what we’re building at Seamless Technologies.