Model Evaluation & MLOps Engineer

Sciforium; San Francisco (CA); On-site; USD 140,000 - 200,000; Full time

Benefits offered by this job

Medical, dental, and vision insurance

401k plan

Daily lunch, snacks, and beverages

Flexible time off

Competitive salary and equity

Job summary

Sciforium is seeking a Research Engineer focused on Model Evaluation & MLOps to build the tools and infrastructure needed to evaluate, deploy, and operate multimodal foundation models reliably. You will enable Sciforium’s models and open-weight models on GPUs, automate quality and performance benchmarking, and improve MLOps workflows connecting research to releases.

Strong Python and ML systems experience are essential, with hands-on work in PyTorch, TensorFlow, or JAX.

Qualifications

• 2+ years of professional ML or software engineering experience in production ML systems or MLOps infrastructure.
• Strong Python and software engineering skills for reliable production systems.
• Hands-on experience with PyTorch, TensorFlow, or JAX and knowledge of modern model architectures.
• Experience with model evaluation, benchmarking, and core lifecycle workflows such as experiment tracking, versioning, deployment, or monitoring.

Responsibilities

• Rapidly integrate new internal and open-weight models into GPU evaluation and inference environments.
• Build automated benchmarks for model quality, latency, throughput, and memory usage.
• Create standardized comparisons across models, baselines, and configurations.
• Build and maintain experiment tracking, model registry, and CI/CD workflows.

Skills

Python

Software engineering

Communication

GPU inference

Experiment tracking

Education

MS or PhD in CS/CE/ML

Tools

PyTorch

TensorFlow

JAX

vLLM

SGLang

TensorRT-LLM

Job description

Sciforium is seeking a Research Engineer focused on Model Evaluation & MLOps to build the tools and infrastructure needed to evaluate, deploy, and operate multimodal foundation models reliably. You will enable Sciforium’s models and open-weight models on GPUs, automate quality and performance benchmarking, and improve MLOps workflows connecting research to releases.

Strong Python and ML systems experience are essential, with hands-on work in PyTorch, TensorFlow, or JAX.