Customer Success
About MyHR
MyHR built a SaaS platform, to make HR straightforward and savvy. Their aspiration is to make HR work for every business where people are taken care of and everyone can focus their energy on building success.
MyHR now supports over 39,322 individual employment relationships in more than 1,680 businesses across New Zealand, Australia and Canada and has a strong further growth aspiration.
Executive Summary
This case study details how a two-phase strategy using data and generative AI platforms revolutionised MyHR support operations.
The consolidation of data sources through the CyberCX Data platform, followed by the implementation of an advanced GenAI platform, enabled MyHR to achieve substantial gains in task automation and workflow efficiency without additional technical support effort, thanks to a secure, scalable, and maintainable solution built with rigorous DevSecOps and IaC practices.
With both platforms built based on AWS-native services, CyberCX were able to deliver a future-proof solution with the highest level of scalability, reliability and access to state-of-the-art Large Language Models. Leveraging reusable, production-ready pattern templates designed and built by CyberCX resulted in significant cost efficiencies for the engagement.
MyHR’s Challenge
MyHR is running a successful SaaS platform however, with increasing demand, their support team started facing increasing operational pressures due to repetitive, manual tasks that consumed valuable time and resources. These tasks spanned multiple systems and required team members to:
- Navigate between disconnected data sources
- Manually extract and reconcile information
- Perform repetitive remediation procedures
- Carry out manual reviews of legally-binding documentation
The organisation needed a comprehensive solution that would augment their support team and automate routine tasks while maintaining security and compliance standards.
Strategic Approach
MyHR relies on a number of heterogenous data sources, so it was decided to organise and democratise the data via Data Platform as a phase 1, and then as a phase 2, add the GenAI platform with the automation of the support team operations workflow.
In early adoption, the GenAI solution handled 605 out of 1,121 cases for the initial use case, saving approximately five minutes per request. This equates to over 50 hours of manual effort saved so far, with a potential to scale to 250+ hours monthly as the solution expands across ~3,000 applicable requests.
Phase 1: CyberCX Data Platform
For the data platform we utilised our battle-tested CyberCX Data Platform accelerator solution, which has been successfully deployed for multiple customers to solve numerous Data and Analytics challenges. The existing Data Platform foundation, such as code templates and the team’s experience with the services required, allowed us to deliver platform in a short period of time.
This phase produced the following outcomes:
- Consolidate disparate data sources into a unified repository
- Standardise data formats and establish governance controls
- Create a single source of truth for operational intelligence
- Build data pipelines for continuous information flow
- Prepare data for downstream consumption by GenAI and other platforms
Phase 2: CyberCX GenAI Platform
With the Data Platform in place, we proceeded to develop a GenAI platform, based on the latest AWS services in Generative AI space such as Bedrock flows etc. The platform was built following the same approach as the CyberCX Data Platform with security, modularity, maintainability and cost-effectiveness in mind.
We utilised AWS’s latest developments in the space of GenAIOps, such as Prompt Management and Bedrock Flow, which allowed us to build a solution that is easy to test and deploy to multiple different accounts (Dev/Test/Prod). Bedrock Flow allowed us to orchestrate complex workflows with multiple different LLM invocations in a maintainable and extendable way.
This phase produced the following outcomes:
- Complex document-processing and evaluating workflows automated through the use of GenAI LLMs and precisely crafted prompt systems. During the initial rollout, the platform automated 605 of 1,121 applicable support tasks—delivering over 50 hours of time savings. With full adoption across the estimated 3,000 monthly requests, the solution is expected to save more than 250 hours of effort each month.
- RAG solution built for the internal MyHR knowledge base which can also be used in other initiatives such as an internal search, customer-facing chatbots and any other use cases where fast access to the large unstructured dataset is required.
- Testing strategy was implemented, which allows individual components of the workflow to be tested, as well as the entire workflow and assert RAG retrieval and LLM response qualities. By doing this, we are ensuring the solution is aligned with principles of Responsible AI. RAGAS tests were also implemented to assure the accuracy of the RAG solution.
The Outcome
Implementation Details
Data Platform Architecture
The data platform implementation featured:
- Cloud-native data lake architecture with multi-tier storage which follows the Modern Data Architecture blueprint.
- ETL/ELT pipelines for automated data ingestion and transformation
- Comprehensive data cataloguing and metadata management
- Role-based access controls and encryption mechanisms
- Extendable design that allows to expand the platform and ingest more data from wide range of sources
GenAI Platform Integration
Building on the data foundation, the GenAI platform included:
- Automated workflow orchestration for common support scenarios
- An automated process for adding information to the knowledge base used by RAG
- All infrastructure secured with data protection at rest and in transit
- Cross-account deployment mechanisms with approval workflows
- Test suit that allows various levels of validation at the component and integration levels
DevSecOps and IaC Implementation
Both platforms were built with security and operational excellence at their core:
- Infrastructure defined and provisioned through code (Pulumi/CDK)
- CI/CD pipelines for automated testing and deployment
- Security controls integrated throughout the development lifecycle
- Comprehensive logging and audit capabilities
What’s next
While building and demonstrating the capabilities of the GenAI Platform and AWS services overall, it was apparent to MyHR, that there are so many other use cases in the business which will benefit tremendously from this new capability.
Use cases such as an internal search tool that will simplify and democratise access to key information and a chat bot which can drive customer engagement at an early stages of their interactions with MyHR etc.
MyHR understands that the adoption of Data and GenAI is not a luxury anymore, but a means to ensure that the company continues to flourish in the continuously evolving technical and business landscape.
Close collaboration between the CyberCX and MyHR engineers allowed knowledge transfer and provided a great opportunity for the MyHR team to gain hands-on experience working with the new technology that is reshaping the future.
CyberCX are excited to see the new ideas and innovations in HR that this engagement will ignite!
“From scattered data to smart automation, CyberCX helped us to build a secure, scalable GenAI platform that frees our HR experts to focus on people—not process.”
Peter Simmons, CTO
MyHR
