AI Architect - Santa Clara, CA - Fulltime

Sourced from dice.com

We are seeking a Senior AI Architect to design and implement production-grade intelligent layers for customer-facing applications. The role focuses on owning model quality, RAG accuracy, prompt engineering, and AI safety across multi-modal systems including Socratic tutors and adaptive learning engines. You will architect complex LLM call chains, optimize latency, and ensure robust safety guardrails before release.

• Architect and own the intelligence layer for multiple programs, including RAG accuracy and AI safety.
• Design and optimize 6-LLM call chain orchestration using NeMoGuardrails for intent classification and synthesis.
• Implement production-grade AI quality metrics, ensuring accuracy thresholds and latency requirements are met.
• Conduct InfoSec adversarial testing and red-teaming to validate safety guardrails before Release 1.
• Build feedback loops into retrieval systems to continuously improve model performance.

• 15+ years of total IT experience with 4-7 years in software engineering.
• At least 2 years focused on LLM application development in production environments (not research or demos).
• Proven track record of shipping LLM-powered features where real users depend on accuracy.
• Expertise in multi-step LLM chain orchestration (LangChain, LlamaIndex) and multi-turn conversation design.
• Deep knowledge of RAG pipeline design, vector similarity search tuning, and hybrid search strategies.
• Strong skills in Python, async programming, and API design for AI services.

• Experience with adaptive learning systems or personalization engines.
• Knowledge graph integration with RAG.
• Multi-agent orchestration patterns.
• ServiceNow API integration experience.
• Prior experience building AI products on NVIDIA infrastructure.