Want to know more?
Are you interested in this project? Or do you have one just like it? Get in touch. We’d love to tell you more about it.
Transforming HR processes through AI at a prominent US research university
Education
How improving data quality, access, and management enabled a leading American research university to harness AI for streamlined HR processes.
University teams faced persistent difficulties; simple HR-related queries were consistently sluggish, with responses often derived from inaccurate or incomplete data accessible only to colleagues having specialized technical skills. This inefficiency was not only detrimental to productivity but also hindered executive decision-making across HR functions.
Recognizing the potential of artificial intelligence (AI) technologies and Large Language Models (LLMs) to enhance decision-making, and drive operational efficiency, the university chose this opportunity to harness AI, addressing a pressing, real-world challenge. The aim was to transform HR operations by improving data accessibility and quality.
In collaboration with Equal Experts, the university took part in a three-month proof of concept to revolutionize HR data management through AI. This initiative involved the implementation of cloud infrastructure enhancements, adherence to data engineering best practices, and the integration of natural language LLM solutions. Together, these efforts helped find and cut information silos, enabling teams to address complex HR queries more efficiently.
The project successfully highlighted how the university could use AI and LLMs to modernize its HR data management processes. Users across the institution gained the ability to extract insights effortlessly and make data-driven decisions more rapidly. Importantly, this initiative not only streamlined current HR operations but also laid a robust foundation for the future adoption of LLM technologies across the university.
This case exemplifies how targeted investments in technology and strategic partnerships can yield substantial operational efficiencies and empower organizations to navigate challenges with greater agility.
Outcomes
3 months
to develop proof of concept LLM
Instant insights
eliminating delays and bottlenecks
AI platform beginnings
a foundation for expanding AI use
About the client
Our client is a leading US university, with a large focus on research. A nonprofit organization, it aims to pursue groundbreaking research while supporting its students through higher education and college life.
Industry
Higher Education/ Not-for-profit
Location
USA
Length of project
3 months
Challenge
Fragmented and Inaccurate Data Hindering Critical Decision Making
Human resources (HR) play a pivotal role in managing personnel across an organization. Teams need accurate HR data to address essential questions like employee retirement eligibility, pending performance reviews, and remaining vacation days for team members.
However, when this crucial information is locked within fragmented, siloed systems characterized by inconsistent definitions and duplicated data, addressing even the simplest inquiries can transform into a lengthy and resource-intensive process. For our client, this situation was a significant opportunity to assess its ability to implement innovative AI technologies while simultaneously addressing a pressing organizational challenge.
Solution
AI-Powered Proof of Concept Focused on Data Best Practices and Natural Language Processing
Collaborating closely with the university’s HR, data, and technology teams, we took an iterative approach to address the challenge within a constrained three-month timeline.
Initially we thoroughly assessed the university’s requirements and data landscape, conducting an in-depth analysis of HR data systems, finding pain points, and defining success criteria. From this first assessment, we systematically addressed the underlying data issues by ingesting raw HR data into a cloud-based Snowflake data warehouse and constructed a robust data model for efficient querying. Data quality was enhanced by standardizing column names and data formats and dropping duplicate records.
With a foundation of quality data, we focused on the LLM component, evaluating and prototyping an AI-driven solution. We explored several LLMs selecting Mistral given its superior performance in this setting. In our work with complex query generation, we discovered that strategic prompt engineering yielded more effective and efficient outcomes. Next, we designed a prompt strategy that enabled the LLMs to generate correct SQL queries from natural language inputs, ending the need for costly fine-tuning. This was seamlessly integrated into a user-friendly interface, allowing non-technical users to query data independently while reducing reliance on technical staff.
Throughout the project, stakeholder engagement was prioritized through rigorous demonstrations and feedback sessions, allowing development that aligned the solution with the real-world needs of end users. Additionally, we implemented infrastructure as code using Terraform to provision cloud infrastructure, ensuring reproducibility and scalability for the future. Comprehensive documentation, including technical guides and architecture diagrams, was also created to support future engineers in building on the proof of concept, thereby securing long-term value for our client.
Results
Enhancing HR efficiency, data quality, and decision-making with a scalable AI solution
The three-month proof of concept successfully confirmed the university’s ability to integrate AI and LLMs as transformative tools for data management and decision-making.
Key outcomes included:
- Enhanced Data Accessibility: Non-technical users can now quickly query HR data using natural language, significantly reducing dependency on technical teams. As a result, valuable technical resources can now focus on initiatives that drive greater business value.
- Improved Data Quality: The standardization and cleansing of data have bolstered the reliability of analyses and reports, leading to more accurate insights.
- Efficient Infrastructure: The adoption of automation and infrastructure as code has streamlined deployment processes, enabling scalable and consistent operations.
- Empowered Decision-Making: Quicker access to correct HR data fosters improved decision-making across the university. Beyond addressing the immediate challenges associated with accessing and using HR data, the project established a robust foundation for AI-driven innovation throughout the institution. By embedding data ingestion, modeling, and cleansing procedures alongside the development of necessary cloud infrastructure, the organization is well-positioned to rapidly extend the LLM solution to more data sets and departments, thus delivering long-term value beyond the first HR data use case.