Member of Technical Staff – Inference Infrastructure

About the role

Join causal’s mission to build a Large Physics foundation Model that can predict and control physical systems. As a Member of Technical Staff focused on Inference Infrastructure, you will enable rapid, large‑scale model evaluation so research never stalls on compute limits.

Key responsibilities

• Build high‑throughput inference systems for large‑scale backtesting, scoring, and evaluation against historical physical observations.
• Design and implement latency, throughput, and efficiency improvements for real‑time inference.
• Optimize the inference stack to fully exploit hardware FLOPs, bandwidth, and memory.
• Extend orchestration frameworks such as Kubernetes, Ray, and Slurm for distributed, large‑batch inference sweeps.
• Establish standards for reliability, observability, and reproducibility across the inference pipeline.
• Collaborate with researchers to support high‑performance inference for emerging model architectures.

Required profile

• Relentless problem‑solver with rapid execution and quick learning in unfamiliar domains.
• Experience building or optimizing inference and serving systems for high throughput and low latency (e.g., TensorRT).
• Strong understanding of distributed compute, GPU parallelism, and hardware‑aware optimization.
• Deep familiarity with deep‑learning frameworks such as PyTorch and JAX.
• Proven ability to write performant, maintainable code and debug complex codebases.

Required skills

• TensorRT
• GPU parallelism
• Distributed compute
• PyTorch
• JAX
• Kubernetes
• Ray
• Slurm
• vLLM (optional)
• SGLang (optional)
• Triton (optional)

What we offer

• Opportunity to work on cutting‑edge AI infrastructure that powers groundbreaking physics research.
• Collaborative environment with experts from DeepMind, Waymo, Cruise, and CERN.
• Competitive compensation and benefits package.