AI Architect - Generative AI
Posted Updated
Summary
Designs enterprise generative-AI architecture — LLMs, RAG, multi-agent orchestration, embeddings, and guardrails — while leading architecture reviews across AI/ML, backend, and data teams. Works with Python/Java, LangChain/LangGraph, AWS Bedrock or Azure OpenAI.
Key Responsibilities
- Design AI architecture, including LLMs, RAG, multi-agent orchestration, embeddings, and model routing.
- Architect AI asset discovery connectors for models, datasets, agents, prompts, and endpoints.
- Implement AI guardrails for PII detection, prompt injection, jailbreaks, and content moderation.
- Lead architecture reviews and guide AI/ML, backend, and data engineering teams.
- Define LLM evaluation, observability, performance, latency, and cost optimization standards.
- Ensure enterprise security, scalability, and alignment with AI governance frameworks.
- Strong programming experience in Python and/or Java.
- Hands-on expertise in GenAI, LLMs, RAG, vector databases, and prompt engineering.
- Experience with LangChain, LangGraph, Semantic Kernel, and familiarity with MCP.
- Experience with AWS Bedrock, SageMaker, Azure OpenAI, or Azure AI Foundry.
- Strong knowledge of APIs, microservices, event-driven architecture, RBAC, SSO, and audit logging.
- Familiarity with LangSmith, Langfuse, Ragas, or Arize.
- 12-25 years