Director, Product AI Platform Architect

Applied Materials • Santa Clara, CA

13h • Onsite

About The Position

We are developing a single AI platform that serves multiple business units, running inside an air-gapped environment. This role owns the architecture and the technical delivery of the platform. You will interface with senior architects and principal AIML Ops engineers, set technical direction, work across business units to turn business problems into common platform capability, and stay hands-on to lead development.

Requirements

• Bachelor’s degree in Computer Science, Computer Engineering, Electrical Engineering or a related field. Master’s preferred.
• 15+ years of experience in software engineering, including 8+ years in architecture roles and 5+ years leading engineering organizations with managers reporting in.
• Demonstrated ownership of a distributed platform architecture at scale, covering domain decomposition, API and event contract management, resilience and consistency strategy, multi-tenancy and tenant isolation.
• Hands-on architecture experience across public cloud (AWS and/or Azure) and Kubernetes or OpenShift, including container packaging, orchestration, GitOps and lifecycle management.
• Experience architecting for both on-premises customer-controlled deployments and cloud-hosted delivery, including the trade-offs between them and the realities of operating under customer security governance.
• Fluency in architecture documentation and governance practice: C4 or equivalent modelling, architecture decision records, design review standards, fitness functions, NFR definition and technical debt management.
• Proven accountability for delivering a platform to committed dates across multiple squads, covering release planning, dependency and risk management, capacity planning and quality ownership. Architecture authority without delivery accountability is not sufficient for this role.
• Direct experience architecting and operating CI/CD pipelines and a production MLOps lifecycle and using engineering metrics such as DORA to diagnose and improve delivery performance.
• Deep hands-on capability across the technology stack, with the credibility to set the reference implementation, review design and code, and lead development efforts directly. This role is not purely advisory.
• Experience running a large multi-stack application portfolio to common engineering standards, and executing migration of legacy systems onto a shared platform.
• Practical experience deploying machine learning or LLM-based capability to production, including GPU capacity planning, inference serving and model optimization for constrained environments.
• Demonstrated ability to elicit and structure requirements directly from business and customer stakeholders through process mapping, current and future state analysis, use-case definition and workflow analysis, and to translate them into domain models, capability maps and architecture decisions.
• Track record of building a shared platform across business units with competing priorities, converging overlapping requirements into common capability and sustaining that alignment when local incentives pull against it.
• Ability to separate genuine platform capability from customer or BU-specific customization, and to design the configuration and extension model that absorbs the difference without forking the codebase.
• Experience building and defending the business case for platform investment in commercial terms such as engineering cost avoided, customer time to value, deployment cycle time and compliance exposure.
• Comfort operating in customer-facing technical settings, including requirements workshops, design reviews, security assessments and qualification discussions.
• Semiconductor industry experience in fab operations, equipment software, process control, yield analysis, manufacturing systems or design enablement, with enough depth to reason about customer process problems and not only their software symptoms. This is a hard requirement.

Nice To Haves

• Direct engagement with leading logic and memory fabs as a software supplier, including participation in security and qualification reviews.
• Experience taking an on-premises product to a multi-tenant SaaS delivery model, including the tenancy, metering, onboarding and operational changes that requires.
• Experience with digital twin, simulation or virtual metrology software in a manufacturing context.
• Familiarity with semiconductor data and equipment integration standards and factory automation interfaces.
• Experience with open-weight model deployment, GPU cluster scheduling and inference optimization frameworks.
• Prior ownership of a reusable component library or internal developer platform adopted across a large application portfolio.
• Experience establishing AI governance covering model risk management, acceptable use policy and evaluation standards in a regulated or IP-sensitive enterprise.
• Background in developer enablement at scale, including technical community building, hackathons, champion networks and measured adoption of engineering tooling.

Responsibilities

• Platform architecture. Own the target-state reference architecture: domain-driven decomposition, bounded contexts, ports and adapters, API-first and event-driven boundaries, documented as C4 views and decision records.
• Deployment flexibility. Architect one system that runs on premises in restricted networks, in private cloud, and as a multi-tenant service in public cloud, with hosting model set by configuration rather than a fork.
• Cloud-native runtime. Set the Kubernetes and microservices standard: Helm packaging, operator patterns, GitOps reconciliation, service mesh where justified, and the scale, caching and sharding strategy for the hosted path.
• Embedded AI. Define the AI architecture for models running on customer hardware and in our cloud: open-weight model selection and quantization, GPU inference serving with continuous batching, model registry and versioning.
• Grounding and guardrails. Own the retrieval-augmented generation design covering embeddings, vector store, hybrid retrieval and reranking, with tool-use patterns that are deterministic, bounded and fully traceable.
• AI assurance. Establish evaluation and model-risk practice: golden datasets, evaluation harnesses, groundedness metrics, drift detection, prompt-injection defenses, and a clear position on build versus adapt.
• Pipeline ownership. Own CI, CD and MLOps end-to-end: trunk-based development, automated testing and quality gates, container scanning and SBOM, progressive delivery, infrastructure as code, model lineage and drift monitoring.
• Technical delivery. Hold the teams to the architecture and to the date. Own release sequencing across squads, dependency and risk management, capacity planning, and the DORA metrics that show delivery health.
• Business to architecture translation. Work with business units and customers to understand process, constraints and pain points, then convert them into domain models, capability roadmaps and architecture decisions.
• Cross-BU leadership. Serve as senior technical interface to product management, field engineering, manufacturing systems, data science, IT and InfoSec. Chair architecture governance and hold one sequenced roadmap.
• Configurability. Own the extension model: plugin and adapter interfaces, versioned APIs with backward compatibility guarantees, declarative configuration, and tenant isolation across silo, bridge and pool patterns.
• Systems integration. Architect integration to equipment and factory automation interfaces, MES, EDA and yield management systems, historians and data lakes, behind stable anti-corruption layers.
• Non-functional ownership. Own SLOs and error budgets, OpenTelemetry observability, resilience patterns such as circuit breaker and bulkhead, zero-trust segmentation, and data governance for residency, retention and egress.
• Hands-on leadership. Stay credible across the technology stack, including Java and Spring Boot, Python, Node.js, .NET, React and Angular, and lead development directly where the risk warrants it.
• Modernization. Own strangler-fig migration of legacy systems onto the common platform, a published deprecation policy, and a funded technical debt register.
• Organization. Interface with managers and principal engineers across different teams and geographical sites.

Benefits

• supportive work culture that encourages you to learn, develop, and grow your career
• comprehensive benefits package
• participation in a bonus and a stock award program, as applicable

What This Job Offers

• Job Type: Full-time
• Career Level: Director