Sr. Director Data & AI Platforms

Responsibilities

• Own and evolve the canonical reference architecture for the Industrial AI platform — spanning data ingestion, processing, model serving, and agentic orchestration layers.
• Define the architecture of the enterprise AI data platform including lakehouse, feature stores, vector databases, streaming pipelines, and real-time inference infrastructure.
• Architect the agent platform: design the orchestration frameworks, tool registries, memory systems, and safety guardrails that enable reliable multi-agent AI workflows at enterprise scale.
• Establish platform layering principles — separating concerns between infrastructure, platform services, AI capabilities, and application-level solutions to ensure modularity and replaceability.
• Drive platform simplification initiatives: consolidate redundant tooling, reduce operational surface area, and establish "golden path" patterns that make building AI applications faster and more reliable.
• Maintain a continuous technology watch across AI platform, data engineering, agent frameworks, and edge computing domains — synthesizing signals from research, open-source, and vendor communities into actionable architectural guidance.
• Lead structured evaluation of emerging technologies (new foundation model APIs, agentic frameworks, vector retrieval architectures, edge AI runtimes, next-gen data formats) using rigorous PoC and architecture fitness criteria.
• Serve as the organization's internal thought leader on platform evolution — publishing architecture decision records, technology briefings, and roadmap recommendations to CoE and enterprise leadership.
• Build relationships with hyperscaler architecture teams, AI platform vendors, and open-source project leads to gain early visibility into emerging capabilities and influence platform direction.
• Identify and mitigate architectural technical debt proactively, proposing migration paths before legacy patterns constrain AI capability delivery.
• Design cloud-native AI platform architectures on major hyperscalers including managed AI/ML services, serverless inference, cloud-native data platforms, and AI gateway patterns.
• Architect for edge and near-edge AI deployment patterns for industrial environments: model compression and optimization for edge hardware, OT/IT integration, edge inference orchestration, and edge-to-cloud data synchronization.
• Define hybrid architecture patterns that span cloud and on-premises — addressing data residency requirements, network latency constraints, air-gapped environments, and operational consistency across deployment tiers.
• Design for industrial-grade reliability: architect patterns for fault tolerance, graceful degradation, offline operation, and deterministic failover in environments where downtime has direct operational consequences.
• Establish FinOps-aligned architecture patterns that balance AI platform capability with cloud cost optimization across training, inference, and data processing workloads.
• Convene and lead the Forge Data and AI Architecture Forum across the enterprise with various product architecture teams and align on standards and changes.
• Define and govern architecture review processes for Data and AI initiatives: establish design review criteria, facilitate reviews, document decisions, and maintain an architecture decision record (ADR) library.
• Partner with solution architects embedded in business domains to translate domain-specific AI requirements into platform capability investments and reusable architecture patterns.
• Drive consistency across the architect community by developing shared pattern libraries, reference implementations, and architecture blueprints that accelerate solution design across the enterprise.
• Represent the Forge AI architecture perspective in enterprise architecture governance bodies, ensuring AI requirements are reflected in enterprise technology standards and roadmaps.

Qualifications

• 10+ years of hands-on architecture experience designing production AI/ML platforms.
• Demonstrated ability to architect end-to-end ML systems: data pipelines, feature engineering, model training, serving, monitoring, and feedback loops at enterprise scale.
• Deep, production-proven expertise with cloud AI and data services on at least one major hyperscaler (AWS SageMaker / Bedrock, Azure ML / OpenAI Service / Fabric, or GCP Vertex AI / BigQuery).
• Ability to architect multi-cloud or cloud-agnostic AI platforms.
• Hands-on architecture experience with large language model platforms and agentic systems, including RAG pipeline design, tool-use frameworks, multi-agent orchestration patterns (LangGraphor equivalent), vector database selection and integration, and LLM inference optimization.
• Proven experience designing hybrid or edge deployment architectures — including at least one industrial or operational technology (OT) environment.
• Understanding of edge inference runtimes, OT/IT network segmentation, data sovereignty constraints, and real-time latency requirements.
• Track record of reducing platform complexity — consolidating toolchains, designing internal developer platforms, establishing golden-path templates, and measurably improving developer productivity and system operability for AI teams.
• Experience leading architecture communities of practice, facilitating architecture review boards, and producing governance artifacts (ADRs, reference architectures, technology radars) that are actively adopted by engineering teams.
• Demonstrated ability to present complex architectural strategies to executive and non-technical audiences, build cross-functional alignment, and influence technology investment decisions at senior levels.
• Strong grounding in modern data architecture: Lakehouse (Delta Lake / Iceberg), streaming platforms (Kafka / Flink / Spark Streaming), data mesh principles, data governance integration, and data quality at scale.
• Deep experience with MLOps platforms (MLflow, Kubeflow, or cloud-native equivalents), including automated retraining pipelines, model governance, drift detection, A/B testing infrastructure, and AI audit trail design.

Preferred Qualifications

• MS or PhD in Computer Science, Machine Learning, Data Engineering, or a related field — or equivalent deep self-directed research and applied experience in AI systems design.
• Familiarity with industrial AI use cases: predictive maintenance, quality inspection, process optimization, supply chain AI, digital twins, or energy management.
• Experience integrating historian data (OSIsoft PI / AVEVA), SCADA, or IIoT platforms is a significant differentiator.
• Knowledge of data security architectures for AI: confidential computing, differential privacy, federated learning, model watermarking, adversarial robustness patterns, and AI-specific access control design.
• Active contributions to open-source AI or data projects, published architecture papers, conference presentations (NeurIPS, Data+AI Summit, KubeCon, re:Invent, etc.), or recognized industry blog authorship in AI platform domains.
• Experience with real-time AI systems: low-latency feature computation, online learning, streaming inference, event-driven AI pipelines, and complex event processing in industrial or financial contexts.
• Experience designing portable AI platforms using abstraction layers (Kubernetes, KServe, Ray, Terraform) that minimize hyperscaler lock-in while leveraging cloud-native capabilities where appropriate.
• Knowledge of responsible AI architecture patterns: explainability infrastructure, bias detection pipelines, human-in-the-loop systems, AI audit logging, regulatory compliance architectures (EU AI Act, ISO 42001).