Senior Machine Learning Platform Engineer
amgen
Job Description
- Design and build reusable ML and GenAI platform capabilities that support model development, experimentation, evaluation, deployment and production operations across multiple teams and use cases.
- Build self-service platform services, APIs and automation that abstract infrastructure complexity and enable developers to provision and consume AI capabilities consistently.
- Develop and maintain MLOps capabilities including experiment tracking, model and prompt registries, evaluation frameworks, deployment workflows and automated promotion across environments.
- Build model-serving and inference capabilities that support classical ML models, deep-learning models and LLMs through scalable REST, gRPC or event-driven interfaces.
- Develop platform capabilities for GenAI and agentic systems, including model access, prompt management, embeddings, vector search, Retrieval-Augmented Generation, tool integration and agent frameworks.
- Engineer integrations with major cloud-based AI and data platforms, using APIs, SDKs and managed services to provide reusable enterprise capabilities.
- Build and maintain containerized platform services using Docker and Kubernetes, including deployment patterns, scaling strategies, service configuration and lifecycle management.
- Design and implement platform APIs, SDKs, templates and shared libraries that establish standardized development patterns and reduce duplication across engineering teams.
- Implement comprehensive observability and operational monitoring, including logs, metrics, distributed tracing, service health, model/LLM usage, latency, errors and operational dashboards.
- Implement AI evaluation and quality-management capabilities, including automated evaluation pipelines, regression testing, model comparison and release-quality gates.
- Build security and governance controls into platform capabilities, including authentication, authorization, secrets management, data access controls, auditability, lineage and responsible-AI controls.
- Design platform mechanisms for usage metering, cost visibility and optimization, enabling teams to understand infrastructure and AI-service consumption.
- Engineer platform services for scalability, reliability and resilience, including retries, asynchronous processing, concurrency controls, fault tolerance and graceful failure handling.
- Develop and improve CI/CD pipelines for AI platform services, reusable components and model-based workloads, including automated testing, artifact management and environment promotion.
- Evaluate new AI, ML and cloud technologies and determine when they should be introduced as shared platform capabilities versus application-specific solutions.
- Partner with data scientists and ML engineers to identify recurring development and operational challenges and convert them into reusable platform patterns and services.
- Provide technical guidance on architecture, scalability, performance, security and production-readiness for ML and GenAI workloads.
- Participate in architecture reviews, code reviews, incident resolution and production troubleshooting across application, platform and infrastructure layers.
- Create and maintain technical designs, architecture decision records, development standards, operational runbooks and platform documentation.
- Help define the longer-term technical roadmap and engineering standards for ML and GenAI platform capabilities.