Staff Engineer - AI

In this Staff Engineer – AI role, you set and evolve the technical direction for AI-enabled products, leading end-to-end AI delivery from concept to production. You will guide architecture, establish best practices, and drive scalable, secure systems while collaborating with cross-functional teams. Your impact includes shaping responsible AI, evaluating trade-offs, and delivering high-quality AI experiences at scale. This is a chance to influence engineering strategy and contribute to a fast-paced, mission-driven tech culture.

• Set the technical direction and roadmap for AI engineering and AI-powered products.
• Lead end-to-end AI delivery from concept to production, including architectural decisions and trade-offs.
• Build production-grade AI systems (LLM-powered apps, RAG, vector search, tool calling, agentic workflows, context management).
• Establish evaluation, observability, tracing, monitoring, and graceful degradation; ensure reliability in production.
• Embed security, safety, governance, and compliance across the AI lifecycle; collaborate with InfoSec, Data Privacy, Legal, and AI Governance.
• Influence engineering strategy across teams; participate in cross-team architectural reviews and decision-making; guide model/vendor selection.
• Communicate trade-offs and risks to both technical and non-technical stakeholders; demonstrate architectural credibility and leadership.

• Hands-on experience building production LLM applications.
• Strong depth in RAG, vector search, tool calling, agentic AI systems, and context management.
• Experience designing evaluation frameworks (offline and online) for AI outputs.
• Ability to navigate model capability, grounding, latency, cost, observability, and reliability in production AI.
• Understanding of AI safety, responsible AI risks, and production safeguards; familiarity with regulatory considerations.
• Judgement and influence expected of a Staff Engineer; ability to lead across teams without formal management authority.

• Strong technical judgment
• Cross-team collaboration
• Clear communication of architectural trade-offs
• LLM applications in production
• RAG (Retrieval-Augmented Generation)
• Vector search