Key Responsibilities:
Solution & Architecture Design
- Own the end-to-end solution architecture for the Agentic AI QA/QC platform, spanning document intake, parsing, validation, formatting, review/approval, and routing workflows.
- Architect the multi-agent framework (e.g., Proofreader, Formatter, Document Converter, Paralegal, Reviewer/Approver, Document Router agents) covering agent orchestration, task decomposition, tool/function calling, and inter-agent communication.
- Define the Retrieval-Augmented Generation (RAG) architecture, including document/knowledge chunking, embeddings, and vector store design for grounding agent responses in SOPs, QC guidelines, and client-specific playbooks.
- Establish standards for agent-to-tool and agent-to-system integration using modern protocols and frameworks such as LangChain / LangGraph and Model Context Protocol (MCP), enabling reusable, composable agent tooling.
Delivery & Technical Leadership
- Provide technical leadership and architectural oversight to AI Engineers, Integration Engineer, and DevOps Engineer through discovery, build, integration, testing, and deployment phases.
- Partner with the Business Analyst and client SMEs to translate business goals and key requirements into technical designs, RBAC models, and integration specifications.
- Guide design and implementation of the document parsing engine, playbook/rules engine, admin/KPI dashboards, and web application UI.
- Support integration design for SharePoint bidirectional sync, ServiceNow connectors, on-prem packaging, and third-party add-in compatibility.
- Drive architecture reviews, technical design sign-offs, and production-readiness assessments across each delivery phase (discovery, platform build, integration, testing, UAT, go-live, hypercare).
- Mentor AI Engineers on best practices for prompt design, agent evaluation, model routing, and responsible AI development.
- Participate in Agile ceremonies (sprint planning, backlog refinement, reviews) and collaborate with the Scrum Master to ensure architectural decisions align with delivery timelines.
Required Skills & Experience
- 8–14 years of overall technology experience, with at least 3–5 years architecting AI/ML, GenAI, or Agentic AI solutions in production environments.
- Strong hands-on experience with Microsoft Azure AI services — Azure OpenAI, Azure AI Foundry (or Azure AI Studio)/AWS and GCP.
- Proven experience designing and building multi-agent / agentic AI systems, including agent orchestration, task planning, and autonomous workflow execution.
- Working knowledge of agent frameworks and orchestration tooling such as LangChain, LangGraph, Semantic Kernel, or equivalent, and familiarity with the Model Context Protocol (MCP) for standardized tool/data integration.
- Solid understanding of Retrieval-Augmented Generation (RAG) architectures, embeddings, and vector databases (e.g., Azure AI Search, Qdrant, Pinecone, or similar).
- Strong programming background in Java and/or Python, with experience building enterprise-grade APIs, microservices, and integration layers.
- Solid grounding in RBAC, SSO/authentication design, audit logging, and security/compliance practices for enterprise platforms.
- Experience with document intelligence / document processing systems (parsing, OCR, formatting/validation engines) is highly desirable.
- Familiarity with responsible AI and AI governance frameworks (e.g., NIST AI RMF) and experience designing audit-ready AI systems.
- Experience working in Agile/Scrum delivery environments on multi-workstream, enterprise consulting engagements.
Preferred / Good to Have
- Prior experience in legal, professional services, or document-heavy compliance-driven industries.
- Experience with Databricks, LLM hosting/routing across multiple providers (OpenAI, Azure OpenAI, Anthropic, etc.), and LLM evaluation frameworks.
- Exposure to conversational AI and future-state Gen AI document intelligence roadmaps.
- Relevant certifications such as Microsoft Certified: Azure AI Engineer Associate / Azure Solutions Architect Expert.
