Senior Staff AI Platform Engineer — GenAI Evaluation

Pinterest Job Advertisements; Palo Alto (CA); Hybrid; USD 209,000 - 429,000; Full time

Job summary

Pinterest seeks a Senior Staff Software Engineer for the AI Platform to shape and scale our unified evaluation system for GenAI and ML workloads. You will partner with engineering, research, product, and safety to improve quality, launch decisions, and trust in AI-powered experiences.

You’ll identify modernization opportunities, define cross-team strategy, and make durable architectural decisions that balance automated metrics, human feedback, and risk.

Qualifications

• 3+ years designing and operating evaluation systems for GenAI and ML
• Strong Python programming and distributed systems fundamentals
• Experience with GenAI evaluation libraries such as AutoEvals, OpenEvals, Ragas
• 5+ years building scalable platforms for distributed experimentation
• Bachelor’s degree in CS/Engineering or equivalent experience

Responsibilities

• Identify modernization opportunities and scale the roadmap for a company-wide evaluation system
• Define cross-team technical strategy and drive broad adoption of shared evaluation capabilities
• Make durable architectural decisions to improve user experience and reliability of the AI/ML evaluation platform
• Establish evaluation methodologies balancing automated metrics, human judgment, safety, and business outcomes
• Lead complex cross-team workstreams, resolve tradeoffs and risks, and deliver impact

Skills

Python programming

Distributed systems

GenAI evaluation

Cross-team collaboration

Education

Bachelor’s degree in CS/Engineering

Tools

AutoEvals

OpenEvals

Ragas

Job description

Pinterest seeks a Senior Staff Software Engineer for the AI Platform to shape and scale our unified evaluation system for GenAI and ML workloads. You will partner with engineering, research, product, and safety to improve quality, launch decisions, and trust in AI-powered experiences.

You’ll identify modernization opportunities, define cross-team strategy, and make durable architectural decisions that balance automated metrics, human feedback, and risk.