Job Specifications
Role Overview
We are seeking an experienced AI Architect to design next‑generation enterprise AI systems with a strong focus on Agentic AI, advanced RAG patterns, multimodal model integration, and secure orchestration frameworks.
This is a highly technical architecture role focused on model integration, inference optimization, safety, workflow orchestration, document ingestion, and enterprise‑grade AI governance.
The architect will define standards for multi-agent orchestration, tool calling, LLMOps, AgentOps, and overall AI architectural governance across business and engineering teams.
Detailed Responsibilities
AI Strategy, Architecture & Governance
Define enterprise AI architecture incorporating Agentic RAG, Graph RAG, multi-agent orchestration, master–agent frameworks, and tool-calling patterns.
Establish architectural guardrails for tool schemas, response grounding, lineage tracking, auditability, and safe tool execution.
Align designs with Mastercard’s governance, security, and Responsible AI frameworks.
AI Solution Architecture
Architect end-to-end AI applications including copilots, autonomous agents, multimodal systems, knowledge-grounded assistants, and agentic marketing solutions.
Integrate open-source and commercial LLMs, Vision-LLMs (e.g., Florence‑2, LLaVA, Qwen‑VL), and diffusion models.
Focus on secure inference, scalability, and efficient orchestration using frameworks like vLLM, TGI, Hugging Face, and Databricks AI stack.
Knowledge Base (PPT/PDF) Ingestion Pipelines
Design ingestion workflows using Unstructured.io, MarkItDown, Apache Tika, PyMuPDF, python‑pptx, etc.
Build pipelines leveraging OCR, layout understanding, and Vision-LLMs to process complex documents, charts, tables, and diagrams.
Govern ingestion and retrieval using Unity Catalog and Databricks Vector Search.
AI Safety, Guardrails & Evaluation
Architect safety frameworks using DeepTeam, NVIDIA NeMo Guardrails, DeepEval, Ragas, and related evaluation tools.
Ensure grounding, policy enforcement, hallucination control, and topic boundary management.
Implement strong logging, traceability, and Responsible AI compliance.
Technical Leadership
Provide direction on AgentOps, RAG frameworks, ingestion pipelines, prompt engineering, and safe tool-calling patterns.
Conduct architectural reviews and guide engineering teams on best practices.
Communicate complex AI architecture concepts to both technical and business stakeholders.
Candidate Profile (All About You)
We are looking for candidates with:
Deep experience in Agentic RAG, Graph RAG, multi-agent systems, and structured tool calling.
Strong hands-on capabilities with LLMs, Vision-LLMs, and diffusion models (integration & inference).
Expertise with Databricks ML, Unity Catalog, MLflow, Vector Search, and enterprise governance workflows.
Experience with PPT/PDF ingestion frameworks, OCR, semantic chunking, and multimodal enrichment.
Ability to collaborate across business, engineering, and governance teams.
Commitment to Responsible AI, security, and compliance-driven architecture.