Performance Evaluation of CPUs, GPUs, and AMD AI Engines on Time-Series Machine Learning Applications (exjobb)

Location:

Ludvika, Dalarna County, Sweden

Job ID:

R0107303

Date Posted:

2025-09-30

Company Name:

HITACHI ENERGY SWEDEN AB

Profession (Job Category):

Administration & Facilities

Job Schedule: 

Part time

Remote:

No

Job Description:

The opportunity: 

We are looking for a master thesis student at our R&D department, would like to join our multinational organization with lots of development opportunities. Our team is dedicated to creating a future where innovative digital technology brings cleaner energy to people around the world. If this opportunity excites you, we encourage you to apply! 

 Time-series machine learning is increasingly critical in domains such as industrial control, predictive maintenance, and anomaly detection. These applications often require low-latency inference and efficient on-device processing. While CPUs and GPUs are widely used for ML workloads, specialized neural processing units (NPUs) such as AMD's AI Engine architecture offer promising acceleration for time-series models. This project aims to benchmark and compare the performance of CPUs, GPUs, and AMD AI Engines across representative time-series ML tasks. 

 

How you’ll make an impact: 

• Design and implement time-series ML models like RNNs, LSTMs, and CNNs to evaluate performance across CPU, GPU, and AI Engine platforms. 

• Establish strong baselines using PyTorch and TensorFlow, helping us understand how traditional frameworks stack up against AI Engine acceleration. 

• Develop and deploy optimized flows using AMD’s AI Engine programming model, Vitis Unified IDE, and Vitis AI toolchain to unlock low-latency, high-efficiency inference. 

• Benchmark key performance metrics including latency, throughput, energy consumption, and memory usage to guide architectural decisions. 

• Analyze trade-offs in accuracy, quantization, and developer productivity to inform real-world deployment strategies. 

• Provide clear guidelines for hardware selection based on workload characteristics, helping teams choose the right compute targets. 

• Evaluate developer effort and time-to-deploy, ensuring our solutions are not only powerful but also practical and scalable. 

• Live Hitachi Energy’s core values of safety and integrity, taking responsibility for your actions while caring for colleagues and the business. 

 Your background: 

• Bachelor’s or Master’s degree in Computer Science, Electrical Engineering, or related field. 

• Proficiency in Python and experience with ML frameworks (e.g., PyTorch, TensorFlow). 

• Familiarity with embedded development and heterogeneous computing architectures. 

• Exposure to AMD Vitis toolchain and AI Engine programming is a plus. 

• Experience with benchmarking, profiling, and performance analysis. 

• Proficiency in English, both written and spoken. 

 

More about us: 

Are you ready for an exciting new challenge? Does the above description sound like you? Applications will be 

reviewed on an ongoing basis, so don’t delay apply today! Recruiting Manager: Lucas Bottura, lucas.bottura@hitachienergy.com and thesis supervisor: Hasan Basri Celebi, hasan-basri.celebi@hitachienergy.com will answer your questions about the position. 

Qualified individuals with a disability may request a reasonable accommodation if you are unable or limited in your ability to use or access the Hitachi Energy career site as a result of your disability. You may request reasonable accommodations by completing a general inquiry form on our website. Please include your contact information and specific details about your required accommodation to support you during the job application process.

This is solely for job seekers with disabilities requiring accessibility assistance or an accommodation in the job application process. Messages left for other purposes will not receive a response.