Seminar Overview
Fixstars engineers introduce key concepts in AI performance engineering and demonstrate how to identify and resolve inefficiencies across training and inference workflows. You’ll learn how to profile hardware usage, uncover bottlenecks, and apply optimizations that deliver measurable gains in speed and cost savings. The session also shares insights from our hands-on evaluation of the latest NVIDIA GPU (B200)—highlighting key considerations to unlock its full AI processing potential beyond its hardware advantages over previous generations.
Through real-world case studies—including a 5x speedup in Llama-3 training and a 50% reduction in inference costs—you’ll see how organizations are leveraging Fixstars AIBooster to improve system performance and ROI. A demo and detailed walkthrough provide practical guidance you can apply to start optimizing your own AI workloads right away.
Speaker
Ryan is a software engineer at Fixstars Solutions specializing in high-performance C++ programming. His work includes building high-throughput, GPU-accelerated image processing pipelines; designing optimization algorithms inspired by quantum annealing; and enhancing computational lithography workflows.
He holds a master’s degree in computer science from Tufts University, a master’s in physics from the California Institute of Technology, and bachelor’s degrees in physics and mathematics from the University of Rochester.
Agenda
- What is AI Performance Engineering?
- Common Practices in AI Performance Engineering
- Practical Optimization Techniques for AI Workloads
- How Fixstars AIBooster Solves These Challenges
- Q&A Session
Total 40 minutes
AIBooster is a performance engineering platform for continuously observing and improving the performance of AI workloads.
Through comprehensive dashboards, users can visualize the utilization efficiency of various hardware resources—including CPU, GPU, interconnect, and storage—as well as identify software bottlenecks to analyze AI workload performance characteristics. Furthermore, by applying optimization frameworks specifically designed for AI workloads, users can achieve efficient performance improvements.
Target Audience
- AI Infrastructure Managers and Engineers
- Technical Leaders of AI Development Teams
- Project Managers aiming for cost-efficient AI projects
- Data Scientists and Researchers interested in performance engineering
Participation fee
Free
