Employee Attrition Using Intel® oneAPI Tools: Developer Spotlight

Success Story of an Intel® Student Ambassador

Employee attrition is a significant concern for businesses due to the high costs associated with rehiring, retraining, loss of productivity, and workflow disruptions. An Intel® Student Ambassador, Ahamed Thaiyub, developed an AI solution called Workforce 360 to predict employee attrition using Intel® oneAPI tools. It integrates three key solutions: resume parsing, predictive analytics, and performance prediction. By efficiently utilizing Intel’s CPU and GPU resources along with AI Tools from Intel, the solution delivers faster and more accurate attrition predictions, helping companies retain valuable employees and make informed decisions to reduce turnover costs.

About the Student Ambassador

As an ambitious engineer and Intel Student Ambassador, Ahamed Thaiyub thrives on using technology to tackle real-world challenges. With a solid foundation from KPR Institute of Engineering and Technology (India), he has gained valuable experience through diverse internships, refining his skills in data analysis, machine learning, and software development. His contributions at Next99 radio station and Mahat Labs Pvt Ltd in India, coupled with his current role as an ML Research Assistant, highlight his dedication to innovation. Noteworthy achievements, including publications in IEEE and recognition in hackathons, underscore his commitment to driving impactful change and advancing the field of engineering. As an Intel Student Ambassador, he is passionate about leveraging Intel's cutting-edge technologies to drive transformative solutions and inspire the next generation of engineers.

Experience as a Student Ambassador

Ahamed’s tenure as an Intel Student Ambassador has been incredibly rewarding and enriching, filled with opportunities to expand my knowledge and skills in Machine Learning (ML), Deep Learning (DL), Large Language Models (LLMs), and CPU computing. Conducting several workshops on these cutting-edge topics has not only solidified his understanding about them but also enhanced his communication and teaching abilities. Initially, his primary interest was in ML, but Intel's advanced technologies and resources broadened his perspective, fostering a deep appreciation for CPU computing and its crucial role in optimizing ML processes.

The most exciting aspect of him being an Intel Student Ambassador is staying at the forefront of technological advancements and being part of a vibrant community of like-minded individuals. The extensive support and resources from Intel have been pivotal in shaping his academic and career aspirations, guiding him towards a holistic approach to computing. This experience has been a journey of continuous learning and growth, significantly influencing his career trajectory.

Challenges Faced

In the employee attrition project undertaken as part of an Intel’s initiative, his project team faced the challenge of optimizing existing solutions for predicting employee turnover, which were primarily tailored for proprietary GPUs. However, their objective was to create a solution that could seamlessly run across different architectures, specifically targeting Intel CPUs. This presented a significant obstacle as it limited the flexibility, efficiency, and accessibility of the project within Intel’s frameworks.

Their motivation to tackle this challenge stemmed from the opportunity to leverage Intel's advanced technologies to develop a more versatile and inclusive platform. By optimizing the employee attrition model to efficiently utilize Intel CPUs, they aimed to enhance performance and broaden its applicability within the Intel ecosystem. Overcoming this architectural limitation became a driving force for the team, guiding them towards innovation and excellence in predictive analytics and employee retention strategies within the realm of Intel's initiatives. 

Advantage of Leveraging Intel® oneAPI Tools

In the employee attrition project, the Student Ambassador tackled the challenge of optimizing the solution for Intel CPUs by leveraging the capabilities of cross-vendor, multi-architecture supported oneAPI tools. These tools, available free of charge from Intel, facilitated the seamless porting and optimization of my application across various hardware architectures. This approach yielded significant performance enhancements, demonstrating the effectiveness of the solution in overcoming architectural limitations. The successful integration of Intel's advanced tools highlights my ability to create versatile and high-performing applications suitable for diverse hardware platforms, ensuring the success of the project. 

Deep Dive Into The Solution

Fig.1: Methodology Tier 1

Fig.2 Methodology Tier 2 and 3 

Project Results

Fig.3: Accuracy Comparison of Different AI Models

oneAPI tools help optimize CPU uptime, ensuring peak efficiency and performance:

 Fig.4: Sample Output 

Intel’s Software Optimizations Used:

• Intel® Extension for Scikit-learn*
• Intel® Distribution of Modin*
• Intel® Optimization for XGBoost* 
• PyTorch* Optimizations from Intel

The complete project is available on GitHub and DevMesh.

Potential Future Enhancements 

Moving forward, the team aims to optimize Workforce 360 using oneAPI tools, refining predictive analytics and bolstering employee retention strategies. Their goal is to provide organizations with insightful HR solutions for informed decision-making in recruitment and retention. By leveraging machine learning and data analysis, they seek to create a platform that not only predicts attrition accurately but also offers actionable insights to mitigate turnover effectively.

They plan to expand Workforce 360's reach across industries and global markets, promoting workforce stability and organizational growth. Through ongoing research and collaboration with Intel and industry partners, they aim to stay ahead of technological advancements, continuously innovating our solution to meet evolving HR challenges. Ultimately, their project strives to reduce turnover costs, enhance productivity, and cultivate positive work environments worldwide.

Next Steps

Sign up to Intel Tiber Developer Cloud today - get started with accelerated development of AI, HPC and edge-computing solutions! Explore the Intel Student Ambassador Program to utilize Intel® oneAPI tools and Intel Tiber Developer Cloud in collaboration with developer communities.

We encourage you to check out the AI, HPC, and Rendering tools in Intel’s oneAPI-powered software portfolio. 

Additional Resources

• Intel oneAPI programming model
• AI/ML Ecosystem Developer Resources
• Intel’s resources for AI development
• Intel’s AI/ML frameworks optimizations
•  

Notices and Disclaimers

Workforce 360, achieved a significant increase in prediction accuracy by 12% and reduced computation time by 18% due to optimizations with Intel® oneAPI.

Configuration used for Workforce 360 includes Intel® Core™ i7 processors on a personal laptop for development and testing, demonstrating the efficiency of our solution even on non-server hardware.
Development tools such as Intel® Extension for Scikit-learn*, Intel® Distribution of Modin*, Intel® Optimization for XGBoost*, and PyTorch Optimizations from Intel were integral to achieving these performance improvements.

Performance varies by use, configuration, and other factors. Learn more at www.Intel.com/PerformanceIndex .

Your costs and results may vary. 

Intel technologies may require enabled hardware, software, or service activation. 

Intel does not control or audit third-party data. You should consult other sources to evaluate accuracy. 

© Intel Corporation. Intel, the Intel logo, and other Intel marks are trademarks of Intel Corporation or its subsidiaries. 

*Other names and brands may be claimed as the property of others.