ML Systems Engineer — Scalable Distributed Training & Inference

OpenTalent; San Francisco (CA); On-site; USD 180,000 - 260,000; Full time

Job summary

Sciforium is seeking a distributed training and inference engineer to build and optimize the ML software stack for large-scale AI workloads. You will work across CUDA/ROCm runtimes to high-level frameworks like JAX and PyTorch to ensure fast, scalable training and serving.

This role emphasizes deep systems engineering, debugging hardware–software interactions, and optimizing performance at every layer of the ML stack, enabling training and deployment of next‑gen LLMs and generative AI models.

Qualifications

• 5+ years of industry experience in ML systems or distributed training.
• Strong programming experience in Python and C++ with distributed systems.
• Familiarity with ML tooling and profiling tools.
• Solid academic background in CS/CE/EE.

Responsibilities

• Maintain and optimize ML libraries and frameworks across environments.
• Own end-to-end ML software stack from drivers to tooling.
• Tune distributed training for scalability and efficiency.
• Profile and optimize performance across multi-node clusters.
• Debug hardware–software interactions and ensure stability.
• Collaborate with research and kernel teams to improve throughput.

Skills

Python

C++

Distributed systems

ML tooling

Education

Bachelor's or Master’s degree in CS/CE/EE

Tools

Nsight

ROCm Profiler

XLA profiler

TPU tools

Job description

Sciforium is seeking a distributed training and inference engineer to build and optimize the ML software stack for large-scale AI workloads. You will work across CUDA/ROCm runtimes to high-level frameworks like JAX and PyTorch to ensure fast, scalable training and serving.

This role emphasizes deep systems engineering, debugging hardware–software interactions, and optimizing performance at every layer of the ML stack, enabling training and deployment of next‑gen LLMs and generative AI models.