Sr. Director, Responsible AI Science and Engineering

Salesforce Salesforce · Enterprise · San Francisco, California - Palo Alto, Washington - Seattle, Washington - Bellevue, CA

Salesforce is seeking a Sr. Director of Responsible AI Science and Engineering to lead the transformation of their Responsible AI (RAI) practice from manual reviews to scalable, AI-powered automated systems. This role involves defining the technical roadmap, architecting production systems for RAI testing and evaluation, establishing scientific rigor, influencing cross-functional strategy, and leading a team of technical experts. The focus is on building robust infrastructure and scientific foundations to ensure trustworthy AI across the enterprise, with a strong emphasis on evaluation science, adversarial testing, and systematic defenses against AI risks.

What you'd actually do

  1. Define the Technical North Star: Develop the 18-24 month roadmap for moving from manual ethics test reviews to automated, AI-powered testing.
  2. Architect Production Systems: Build robust infrastructure (e.g., automated intake pipelines, LLM-judge patterns, LLM Council approaches) that can handle Salesforce-scale traffic.
  3. Set Scientific Rigor: Establish the research foundations, evaluation metrics, and technical logic for how we test and align AI systems.
  4. Influence Cross-Functional Strategy: Partner with technical and product teams to embed RAI into the development lifecycle across business units.
  5. People Leadership: Act as a "player-coach," providing technical mentorship, managing performance, and unblocking complex challenges in study design, data analysis, and software engineering.

Skills

Required

  • 10+ years in software engineering and/or AI research
  • proven ability to build production-scale systems
  • Hands-on proficiency in Python, SQL, CI/CD pipelines
  • Applied knowledge of Data Science
  • Experience building LLM-powered systems (agent frameworks, RAG, multi-model orchestration)
  • deep knowledge of modern AI architectures
  • Proven track record in evaluation science, adversarial testing (red teaming), and designing systematic defenses against bias, hallucination, and other RAI risks
  • 5+ years managing technical teams (engineers, researchers, data scientists)
  • Strong performance management and operational excellence
  • Demonstrated ability to translate complex technical/ethical narratives into strategic business goals for executive stakeholders
  • Comfort charting direction in high-ambiguity environments, using data to make pragmatic tradeoffs between ambition and deliverability
  • A related technical degree required

Nice to have

  • AI powered coding tools (e.g., Claude Code, Cursor)

What the JD emphasized

  • transforming our RAI practice from manual expert reviews to operating scalable, AI-powered automated systems
  • own the technical architecture, scientific rationale, and operational scaling of our RAI systems
  • leading a team of technical Responsible AI experts (data science, research science, engineering) through a critical transformation
  • proven track record in evaluation science, adversarial testing (red teaming), and designing systematic defenses against bias, hallucination, and other RAI risks
  • leading a career-defining shift in how a major enterprise practices ethics, with the mandate, resources, and executive support to succeed

Other signals

  • transforming our RAI practice from manual expert reviews to operating scalable, AI-powered automated systems
  • own the technical architecture, scientific rationale, and operational scaling of our RAI systems
  • leading a team of technical Responsible AI experts (data science, research science, engineering) through a critical transformation
  • leading a career-defining shift in how a major enterprise practices ethics, with the mandate, resources, and executive support to succeed