Principal Data Architect

Altimetrik delivers outcomes for our clients by rapidly enabling digital business & culture and infuse speed and agility into enterprise technology and connected solutions. We are practitioners of end-to-end business and technology transformation. We tap into an organization’s technology, people, and assets to fuel fast, meaningful results for global enterprise customers across financial services, payments, retail, automotive, healthcare, manufacturing, and other industries.

Founded in 2012 and with offices across the globe, Altimetrik makes industries, leaders and Fortune 500 companies more agile, empowered and successful.

Altimetrik helps get companies get “unstuck”. We’re a technology company that lives organizations a process and context to solve problems in unconventional ways. We’re a catalyst for organization’s talent and technology, helping teams push boundaries and challenge traditional approaches. We make delivery more bold, efficient, collaborative and even more enjoyable.

Position Title:
Principal Technical Architect – AI Systems, Data Platforms & Cyber Security

Department:
Enterprise Architecture / Data, AI & Security Engineering

We are seeking a visionary and hands-on Principal Technical Architect to lead the architecture, design, security, and strategic evolution of our Enterprise Data Platforms and Multi-Agent GenAI Systems. In this role, you will bridge the gap between complex enterprise data engineering, modern cloud architecture, cutting-edge Generative AI applications, and enterprise cybersecurity controls.

You will design zero-touch automated data observability solutions, multi-agent AI pipelines, zero-trust data access patterns, and enterprise-wide GenAI adoption frameworks. The ideal candidate brings a deep technical background in Databricks, cloud platforms, Python, contract-driven LLM architectures, threat modeling for AI systems, and proven enterprise leadership in scaling and securing AI solutions across the organization.

• Architect Multi-Agent AI Pipelines:
  Design end-to-end, LLM-powered multi-agent frameworks (deterministic + generative) using Pydantic contracts, asynchronous Python, and provider-agnostic model integration (e.g., OpenAI SDK, Databricks Model Serving).
• AI Tooling & Scaffolding:
  Build graph-based execution builders, dynamic YAML rules engines, capability registry patterns, and structured diagnostic retry mechanisms for LLM agents.
• Human-in-the-Loop Integration:
  Implement validation and curation layers that enable user DAG editing, schema validation, and error repair before scaffold generation.
• LLM Threat Modeling & Abuse Prevention:
  Lead abuse-case modeling, prompt injection defense, jailbreak mitigation, and red-teaming strategies for LLM agents and RAG architectures.
• Output Guardrails & Data Protection:
  Implement payload masking, custom SQL validation layers, row-count caps, and automated PII/PCI detection to prevent data exfiltration via AI interfaces.
• Identity & Access Governance:
  Architect hybrid identity flows (OAuth 2.0, Okta/Entra ), Service Principal access patterns, and automated token lifecycle/rotation management (e.g., Delta Sharing tokens).

• Databricks Estate Architecture:
  Lead large-scale data platform migrations, estate auto-discovery, and governance automation across Databricks work spaces (Unity Catalog, Workflows, Jobs API, Delta Lake, Delta Sharing).
• Data Governance & Zero-Trust Access:
  Enforce fine-grained authorization models including Row-Level Security (RLS), dynamic column masking, and centralized data classification in Unity Catalog.
• Data Quality & Observability Frameworks:
  Design automated, multi-tiered data quality verification platforms capable of…