Transforming Global Background Checks with DevOps on Google Cloud
Efficiency and Accuracy with AI-Powered Background Checks
19.04.2024

If you have ever been formally employed or have hired people, you know how vital background checks are, and how their speedy, accurate execution makes life easier for employers and employees alike. One of our clients, a Middle East and Africa (MEA)-based background check company with a global reach spanning 200+ countries, specializes in verifying credentials for medical staff, doctors, and other professionals across diverse industries.


Manual processes: Time-consuming, non-scalable, hampering growth


Before they approached us, the client company relied on manual processes for conducting background checks, depending on verifying information across 1,500 known data sources. As you can imagine, this was a time-consuming and error-prone approach. Further, it was also restricted by the number of employees they had, thus stifling their growth potential. There was a critical need for an entirely different way of doing things - one which enabled them to scale up while ensuring high accuracy, speed, and quality.


Going digital: The way ahead


While migrating from a manual process to a digital workflow felt like the obvious answer, it needed careful planning to not only utilize the optimum set of technologies, but also to weave them together into an efficient, modular solution. After a thorough analysis, we decided to leverage DevOps functionalities on Google Cloud to create a fully digital, AI-driven background check system.


Putting the pieces together


When designing any solution, we follow the below guidelines:


1. Scalability - The solution needed to be able to scale up and down as per the volume of work.


2. Automated delivery - Updates and changes to workflows needed to be made available to end-users in a seamless manner with minimal manual intervention.


3. Efficiency - Like in any cloud-based solution, the resource utilization needs to be kept at a minimum to achieve cost efficiencies.


4. AI integration - Replacing human labour with AI was the essence of achieving speed and accuracy in summarization and verification of information.


5. Simplified monitoring - No solution is complete if you are unable to easily monitor it and tweak it if it is not performing to pre-defined standards.


With these goals in mind, this is how we implemented the final solution:


Scalability with Kubernetes


Set up a Google Kubernetes Engine (GKE) cluster using Terraform.Configured Horizontal Pod Autoscaling (HPA), Vertical Pod Autoscaling (VPA), and Node Autoscaling to efficiently handle fluctuating workloads.


  • Implemented custom scaling rules based on document parsing workload, ensuring optimal resource allocation.

  • CI/CD with GitLab


  • Built a robust CI/CD pipeline using GitLab CI, enabling continuous integration and delivery of application updates.

  • Incorporated automated testing in the CI/CD process to ensure high-quality, bug-free releases.

  • Utilized unit tests to verify core functionality.

  • Conducted integration tests to validate interactions between services.

  • Performed end-to-end (E2E) testing to simulate real-world scenarios.

  • Automated the testing process with tools like Selenium and Postman for API validation, ensuring every release met stringent quality standards.

  • Backend optimization for responsive applications


  • Streamlined traffic management with NGINX for backend routing.

  • Ensured rapid data retrieval with Redis clusters for caching parsed data from multiple formats (PDFs via OCR, Word documents, Excel files, etc.).

  • AI and NLP integration for data processing


  • Leveraged Google Cloud NLP services for advanced data summarization, extracting actionable insights from unstructured data sources.


    Simplified scheduling and monitoring


  • Set up Cloud Run for job scheduling, automating background check processes at scale.

  • Integrated Prometheus and Grafana dashboards for real-time monitoring, coupled with ELK Stack + Filebeat for comprehensive logging and alerting.

  • Efficient messaging and data exchange


  • Deployed a Kafka cluster for seamless message and data exchange between microservices.

  • Utilized a clustered MongoDB database to handle large-scale data storage with high availability and reliability.

  • When results speak for themselves


    As the saying goes, the proof of the pudding is in the eating. We're happy to confirm that this pudding was truly delicious. Here’s what we accomplished:


  • 95% reduction in manual efforts - Fully automated background checks eliminated manual intervention.

  • Enhanced speed and accuracy - Background checks were completed faster and with greater precision.

  • Improved quality assurance - Automated testing in CI/CD pipelines ensured error-free, high-quality releases.

  • Seamless scalability - The system scaled up and down effortlessly to meet fluctuating workloads, thanks to Kubernetes and Redis.

  • Improved data insights - AI-driven summarization provided actionable intelligence for faster decision-making.

  • 24/7 system reliability - Continuous monitoring and proactive scaling ensured high availability.

  • Why Choose Truminds for DevOps?


    Put simply, our comprehensive DevOps and SRE approach frees your team up from having to manage infrastructure and lets them focus on innovation and customer satisfaction. Here's what sets us apart:


  • End-to-end responsibility: From cloud infrastructure to CI/CD pipelines, testing, and monitoring, we take care of it all.

  • Flexibility and scalability: Quickly scale or adapt your team to meet changing technology demands.

  • Proactive updates: Stay ahead with our up-to-date tools, technologies, and certifications.

  • Industry expertise: Leverage learnings from other customers and best practices to ensure excellence.

  • Rapid response: Easily augment the team during spikes in activity or emergencies.

  • In addition to the points mentioned, we also offer a seamless single-click deployment process for infrastructure management using Terraform scripts. With our pre-configured Terraform resources, we can deploy and manage our infrastructure effortlessly, ensuring consistency, speed, and reliability. This capability further streamlines operations, empowering our team to focus on delivering value rather than getting bogged down in manual configurations or troubleshooting.

    Like what you see? Let’s Start Something Great!!