A multi portal recruitment platform syncs nationwide job postings across nine external boards

PROJECT OVERVIEW

ABOUT THE CLIENT

This case study is based on a real VNEXT project. The client's identity and certain project details have been anonymized or generalized for confidentiality purposes.

The customer is a recruitment company that helps employers across the country find staff and keep them. It belongs to the world of professional services, where a firm's reputation rests on the quality of its matches, and where reach matters too: the more places a job is seen, the more likely the right person finds it. That thinking shaped the whole product, which had to serve the candidate looking for work, the company hiring, and the staff running the service in between.

THE CHALLENGE

Building the platform meant delivering three distinct environments within a single unified architecture. Candidates required intuitive job search, bookmarking, and direct employer messaging; hiring companies needed applicant tracking and interview pipelines; and the internal operations team required comprehensive control over listings, contracts, and access permissions. While each portal served a different audience with separate workflows, all three relied on a single synchronized data source.

elivering that seamless experience involved several demanding technical requirements. The search engine had to support multi-criteria filtering by region, role, salary, and employment type, along with an automated railway station locator. Engagement had to be immediate, combining real-time messaging with an automated interview-booking chatbot. Beyond its own perimeter, the platform needed to distribute listings across nine external job boards daily and ingest incoming applications back into the central pipeline.

Adding to the complexity, the functional specifications were still shifting as development kicked off. Because our team was responsible for detailed design, implementation, and unit testing, any unresolved requirement or edge case left unaddressed would carry directly into the codebase and risk production stability. 

PROJECT OBJECTIVE

The customer wanted a job matching platform that connects candidates and companies nationwide, and widens the audience for every job by posting it to external boards automatically, then bringing the resulting applicants back into one place where the conversation and the interview can happen.

APPROACH AND METHODOLOGY

Reading the specification closely and asking early

Our team read the customer's specification carefully and raised questions until each requirement was clear. Because the customer was open to those questions and helped answer them, gaps were closed in conversation rather than guessed at in code.

Writing the detailed design so the customer can check it

During detailed design, we wrote as thoroughly as we could and sent it to the customer for review, so a misunderstanding would surface on paper. Technical risks were prepared for in advance, and when a problem needed outside support, our team raised it early so that help could arrive in time.

Keeping an eye on everyone through the daily meeting

The daily meeting was used to follow how each person was getting on, which made it easier to spot where someone needed help and to offer it quickly.

Code was built and deployed through an automated pipeline: a push to GitHub triggers a workflow that builds a Docker image, stores it in the AWS container registry, and deploys it to the managed container service. It is the kind of DevOps discipline that keeps a development environment predictable from one deployment to the next.

Limiting who can see what

Project documents were shared through the customer's cloud drive with view, comment, or edit rights given only to people with a role in the project. Detailed design discussions with the customer ran through a version control system with the same restrictions, and source code lived in the customer's own GitHub account protected by two step verification. Details of the cloud services were visible only to the project manager and technical leader, and only to the extent the customer allowed.

TEAM SCALE

Around eighteen people worked on the project under a project based engagement, spread across management, backend, frontend, testing, and communication with the customer.

• Project managers and a technical leader: ran the delivery and held the technical direction
• Backend engineers: built the Laravel services, the integrations, and the infrastructure on AWS
• Frontend engineers: built the three sites in Vue 3 and TypeScript
• Testers: verified the platform's many flows, drawing on our software testing practice
• Bridge communicators: kept requirements and questions moving between the customer and the team

THE SOLUTION AND TECHNOLOGY STACK

Our team delivered the platform as a web application with three sites and a layer that connects it to the wider job market.

Candidate site

• Job search: filters by area, occupation, the nearest train station, desired salary, working style, and newest listings
• Saved jobs and history: lets candidates keep jobs for later and look back at the jobs they have viewed
• Chat and interview booking: lets candidates talk to a company in real time after applying, with a chatbot that books the interview

Company site

• Applicant and interview management: lets companies confirm interview times and mark candidates as passed or not passed
• Automatic reminders: emails a company when a candidate has applied and received no reply after four days
• Job and billing information: lets company managers see their jobs, applicants, and the amounts payable to the service

Admin site

• Management of jobs, companies, and candidates: adds, edits, deletes, and searches records, with CSV import and export
• Contracts, admin accounts, and permissions: manages agreements with companies and controls what each administrator can do

External job boards

• Daily job feeds: sends a CSV file every day to the FTP or SFTP servers of nine outside job boards so jobs appear there automatically
• Applications coming back in: when someone applies on one of those boards, the platform stores the application, then books an interview and opens a chat as for any other candidate

Technology Stack

The same attention to search and live conversation shaped our work on a healthcare recruitment portal, and the AWS foundation here is part of our wider cloud application development work.

THE RESULTS

• Delivered on the date promised: the platform was completed by the deadline agreed with the customer at the start.
• Accepted by the customer: the customer was pleased with the product it received.
• One place for every applicant: people who apply through the customer's own site and through nine outside boards all arrive in the same system, with the same chat and interview booking.
• A wider audience for every job: daily feeds put each listing in front of candidates on other boards without anyone posting it by hand.
• A smooth handover: all documents and project information were passed to the customer, so the next stage could begin from a full record.

What the customer has now is a recruitment platform that works at the speed candidates expect and reaches well beyond its own site.