Senior AI/ML Software Engineer — Frontier Data Pipelines

Front Door Defense; San Francisco, Northern (CA, KY); Hybrid; USD 208,000 - 315,000; Full time

Job summary

Snorkel AI in San Francisco, CA, is hiring a Senior | Staff Software Engineer for AI/ML to build scalable evaluation pipelines and accelerate frontier-grade data generation at scale. Hybrid work in SF. You will explore how frontier-grade data is produced and evaluated, shape standards, and scale the team’s impact.

The role focuses on efficient evals, routing, and production-ready ML infrastructure with strong emphasis on data-driven quality and cost control.

Qualifications

• 5+ years building production ML or software systems with end-to-end ownership
• Hands-on experience running LLM/ML workloads in production
• Statistics and experimentation: experiment design, hypothesis testing, sampling, confidence intervals
• Strong Python and software engineering fundamentals
• Experience designing evaluations and interpreting results rigorously
• Clear communication with researchers, engineers, and business partners

Responsibilities

• Efficient agentic evals with adaptive sampling and early stopping
• AI model routing to cheapest model that meets quality bar
• Fine-tune and serve open-weight models where they match frontier quality
• Predictive difficulty estimates for frontier tasks
• Build golden datasets and measure calibration of LLM-as-judge systems
• Turn research prototypes into reusable production components

Skills

5+ years building production ML apps

Production ML workloads

Statistics & experimentation

Python & software engineering

Evaluations design & interpretation

Communication with researchers/enginee

Education

MS or PhD in CS/ML/Statistics

Tools

LoRA

Model routing

Open-weight models

Job description

Snorkel AI in San Francisco, CA, is hiring a Senior | Staff Software Engineer for AI/ML to build scalable evaluation pipelines and accelerate frontier-grade data generation at scale. Hybrid work in SF. You will explore how frontier-grade data is produced and evaluated, shape standards, and scale the team’s impact.

The role focuses on efficient evals, routing, and production-ready ML infrastructure with strong emphasis on data-driven quality and cost control.