Staff Product Manager, AI Evaluations

October 13

Apply Now

Receive Emails with Similar Jobs

Dropbox

WebsiteLinkedInAll Job Openings

Cloud Storage • Enterprise • Productivity

Dropbox is a cloud-based service that provides tools for storing, sharing, and accessing files across devices. It offers features such as document sharing, video review, automatic backups, and AI-driven scheduling. Dropbox also provides solutions for different sectors like teams, sales, marketing, and education, and industries including construction, media, technology, and manufacturing. With a focus on security, Dropbox ensures files are encrypted and protected against tampering. It offers integrations with various productivity tools and is trusted by major companies for efficient file management and collaboration.

1001 - 5000 employees

Founded 2007

🏢 Enterprise

⚡ Productivity

📋 Description

• Define and drive the roadmap for Dropbox’s AI Evaluation Framework, covering both quantitative metrics and human-in-the-loop systems. • Partner with ML, data, and research teams to build evaluation pipelines that measure model accuracy, helpfulness, bias, latency, and hallucination rates. • Translate business and product goals into measurable AI performance objectives. • Develop taxonomies and methodologies for evaluating large language model (LLM) behaviors in production. • Work closely with PMs across Dash to integrate eval signals into product development loops. • Partner with infra and AI platform teams to operationalize eval tooling and dashboards. • Ensure evaluations are ethical, transparent, and aligned with Dropbox’s privacy and trust standards. • Communicate findings and drive accountability across teams for improving AI quality.

🎯 Requirements

• 10+ years of product management experience, ideally in AI/ML systems, data platforms, or experimentation. • Proven experience defining and shipping metrics-driven AI products or frameworks. • Strong understanding of model evaluation concepts such as precision/recall, win-rate testing, and user feedback loops. • Excellent analytical, problem-solving, and communication skills. • Strong cross-functional collaboration across ML, data, design, and research teams. • Comfort balancing experimentation speed with rigor and reproducibility.