Sr Applied ML Engineer – Physics-Driven Systems & Optimization

As a Senior Applied Machine Learning Engineer, you will design, implement, and deploy state-of-the-art ML architectures that merge physics insights, numerical optimization, and modern AI techniques.

You'll contribute to building scalable and explainable ML systems, from geometry-aware GNNs and Transformers to reinforcement learning and generative models, that drive design automation, anomaly detection, and optimization in Keysight's next-generation platforms.

• Partner with Keysight experts in RF, EM, circuit, and measurement domains to translate physical constraints and design workflows into ML-ready formulations.
• Design and implement advanced ML architectures:
  • Graph Neural Networks (GNNs) for geometry/topology-aware modeling
  • Transformers for sequential and multimodal data
  • Vision Models (CNNs, ViTs) for field- or spectrogram-based detection
  • Generative Models (GANs, Diffusion) for data augmentation and design candidate generation
• Apply advanced optimization and control methods:
  • Bayesian, gradient-based, and gradient-free optimization
  • Reinforcement Learning (PPO, DDPG, SAC) for continuous tuning and control tasks
• Develop scalable training and inference pipelines (multi-GPU, HPC, AWS) ensuring efficiency and reliability.
• Write production-ready code in Python, C++, and CUDA, integrating with CI/CD pipelines and performance profiling tools.
• Benchmark ML and RL models against physics simulators and measurement datasets for robustness and reproducibility.
• Collaborate with product teams to embed AI/ML-based optimization and generative modules into Keysight software.
• Stay current with the latest ML, RL, and generative AI research; evaluate and prototype promising new techniques.

• Master's or PhD in Applied Mathematics, Scientific Computing, Computer Science, Electrical Engineering, or related field

• 5+ years of experience applying scientific computing and optimization to real-world problems (e.g., RF, EM, or measurement systems)

• Strong hands-on experience with modern ML architectures (GNNs, Transformers, Vision Models, Neural Operators)

• Practical experience with generative models (GANs, VAEs, Diffusion)

• Background in Bayesian and numerical optimization and hyperparameter tuning

• Applied experience with reinforcement learning (PPO, DDPG, SAC)

• Proficiency in Python, C++, CUDA, and GPU performance optimization

• Experience with multi-GPU/distributed training in HPC or cloud (Slurm, MPI, AWS)

• Solid software-engineering discipline (testing, CI/CD, modular design)

• Excellent communication and collaboration skills across cross-functional teams