Staff ML Engineer – Production-Ready Large-Model Systems

Sarsfield Group; United States; On-site; USD 200,000 - 310,000; Full time

Benefits offered by this job

Health and Dental

Life Insurance

401(k) matching

Equity

Job summary

Sarsfield Group is building a production-grade ML platform and seeks a senior ML engineer to own end-to-end systems—from data and training to evaluation, inference, and deployment. You will design and refine training and fine-tuning pipelines for large models while creating robust evaluation mechanisms and scalable data pipelines.

You will architect high-performance inference systems, optimize latency, GPU utilization, and memory, and partner with research and application teams to translate

Qualifications

• Strong software engineering and production-grade code
• Experience building production ML systems
• Knowledge of large-model training and evaluation
• Experience with GPU-based training and inference
• Ability to ship ML features end-to-end
• Strong problem-solving and independent judgment

Responsibilities

• Own end-to-end ML systems from data and training to evaluation, inference, and deployment.
• Build and evolve training and fine-tuning pipelines for large models.
• Design evaluation systems measuring capability, robustness, safety, and product performance.
• Architect high-performance inference systems optimizing latency, GPU utilization, memory, and cost.
• Build data pipelines and reliable production infrastructure for models.

Skills

Python

PyTorch

JAX

GPU training

Production-grade code

System design

Job description

Sarsfield Group is building a production-grade ML platform and seeks a senior ML engineer to own end-to-end systems—from data and training to evaluation, inference, and deployment. You will design and refine training and fine-tuning pipelines for large models while creating robust evaluation mechanisms and scalable data pipelines.

You will architect high-performance inference systems, optimize latency, GPU utilization, and memory, and partner with research and application teams to translate