Senior Agentic AI Solution Architect

Key Responsibilities: Architecture Ownership

• Own end-to-end architecture: ingestion, knowledge/retrieval, agent reasoning, tool access, and governance.
• Design the knowledge lake across telemetry, voice-of-customer signals, CRM, backlog systems, documents, and repositories.
• Architect the MCP gateway, registry, servers, and connectors for governed tool and data access.

Multi-Agent System Design

• Design the multi-agent reasoning core: planner/re-planner, reasoning loops, memory, and agent collaboration.
• Define orchestration, planning, reasoning, and autonomous workflow architectures.
• Architect solutions utilizing LLMs, RAG, knowledge systems, vector databases, and semantic search.

Governance & Evaluation

• Build governance as a first-class layer: audit trails, grounding, guardrails, human-in-the-loop, and graceful degradation.
• Establish AI governance, security, explainability, observability, and Responsible AI standards.
• Define the evaluation harness, quality metrics, and measurable baselines.

Delivery & Technical Direction

• Drive an end-to-end pilot on a real use case before scaling.
• Set technical direction for the offshore delivery pod.
• Create architecture blueprints, reference implementations, and reusable frameworks; guide engineering teams through implementation and technology decisions.

Required Skills & Experience

• 10+ years in software engineering/architecture, including 3+ years hands-on with agentic or multi- agent LLM systems in production.
• Proven greenfield architecture ownership on a complex, ambiguous system.
• Deep multi-agent orchestration expertise: planning, reasoning loops, tool use, memory, and agent collaboration.
• Strong RAG and knowledge architecture: embedding strategy, hybrid/semantic retrieval, vector stores, and grounding.
• Hands-on experience with agentic frameworks (LangGraph, LangChain, or equivalent) and MCP or comparable tool-access layers.
• Demonstrated AI safety and evaluation work: hallucination mitigation, guardrails, and evaluation harnesses.
• Strong Python and/or Java development experience, with API-first and microservices-based architecture background.
• Exceptional client-facing communication — able to hold your own with a senior technical executive.
• Still hands-on: writes code, builds prototypes, debugs real systems.

Preferred / Good to Have

• Regulated or compliance-sensitive domain experience (privacy, GRC, security, financial services, healthcare).
• Consulting or client-embedded background.
• Experience with distributed onshore/offshore delivery models.
• Experience with cloud AI platforms such as Azure AI Foundry, AWS Bedrock, or Google Vertex AI.
• Familiarity with MLOps/LLMOps practices and model evaluation frameworks.

Education

Bachelor’s or master’s degree in Computer Science, Engineering, Artificial Intelligence, or a related field (or equivalent practical experience).