No Green Card :: 14+ years :: 100% Remote :: AI Architect - Azure & Healthcare ::

[Suvendu Adhikari]

Position:  AI Architect -  Azure & Healthcare
Location: 100% Remote
Duration : 12 months
RATE :  $60/hr C2C

• Own the end-to-end architecture of the Agentic AI platform: agent graph design, state management, tool integration patterns, LLM integration strategy, and observability framework.
• Define engineering standards, design patterns, and guardrails for agent development, LLM usage, and platform security.
• Lead architecture reviews and ensure all implementation decisions align with the approved design.
• Collaborate with client enterprise architects, clinical SMEs, and infrastructure teams to validate architectural decisions against operational and regulatory requirements.
• Drive AI governance and review board preparation: model lineage, explainability framework, confidence scoring design, and audit trail architecture.
• Evaluate and select frameworks, tools, and Azure services; define the migration strategy from legacy infrastructure to Azure Cloud.
• Identify and mitigate architectural risks throughout the delivery lifecycle.
Required Skills & Experience
• 10+ years software engineering; 4+ years architecting production AI/ML or LLM platforms.
• Agentic AI architecture: deep expertise in LangGraph, LangChain, or equivalent; state graph design, deterministic orchestration, checkpointing, and HITL patterns.
• LLM platform design: enterprise LLM gateway integration (Azure OpenAI / UAIS), prompt engineering strategy, token economics, confidence scoring, and content safety.
• Azure architecture: AKS, PostgreSQL, Event Hub, Key Vault, API Gateway, Container Registry; cloud-native design patterns and security models.
• Demonstrated experience producing architecture documents, ADRs, and technical governance artifacts for enterprise engagements.
• Experience with AI governance, regulatory compliance (HIPAA or equivalent), and explainability requirements in regulated industries.
Preferred Skills
• Healthcare technology: clinical workflows, prior authorization, claims data, or clinical coding systems.
• LLM observability: Arize AX, LangSmith, or equivalent.
• Experience presenting architecture to executive stakeholders and AI review boards