This list contains only the countries for which job offers have been published in the selected language (e.g., in the French version, only job offers written in French are displayed, and in the English version, only those in English).
Amgen is hiring a hands-on Machine Learning Engineer (Level 4) to design, build and operationalize generative and agentic AI systems that are safe, auditable and production ready. You will own LLM and agent capabilities, evaluation frameworks and guardrails, and work with data science, product, platform and compliance partners to move models from prototype into trusted enterprise services.
Job Responsibility:
Design, implement and deploy LLM solutions and agent frameworks for production workflows, focusing on reliability, interpretability and safety
Build and run evaluation pipelines: automated metrics, unit and integration test suites, adversarial/red-team tests, and human-in-the-loop evaluation
Create and operationalize guardrails: prompt patterns, policy enforcement, input sanitization, output validation, data filtering and explainability tools
Prototype and productionize agentic behaviours: RAGs, multi-step planners, tool-use interfaces, state management and safe action execution
Implement MLOps best practices: model versioning, CI/CD for models, scalable serving, observability and cost controls
Partner with security, privacy, regulatory and ethics teams to embed compliance and auditability into deployments
Stay updated with the latest trends and advancements
Requirements:
4 to 7 years of applied machine learning or software engineering experience, including at least 2 years on production ML systems
B.Tech, M.Tech or MS with specialization in Computer Science
Strong hands-on experience with large language models: fine-tuning, instruction tuning, retrieval augmentation and embeddings
Practical experience with agent frameworks and multi-step agent behaviours
Solid software engineering skills in Python and deep learning frameworks such as PyTorch, JAX or TensorFlow
Experience building evaluation suites, human-in-the-loop workflows and adversarial testing
Demonstrated implementation of guardrails and safety mechanisms for NLP/LLM systems
Experience deploying services in cloud environments and with container orchestration (Kubernetes) and model serving technologies
Clear communicator able to translate prototypes into production solutions and to engage cross-functional stakeholders
Nice to have:
Prior work on agentic AI, multi-agent systems, planners or applied agent safety research
Experience with benchmarking and continuous evaluation tooling
Background in regulated industries such as pharma or healthcare
Familiarity with Traditional ML, data governance and model risk management
Advanced technical credentialing or proven contributions to open-source or academic work in LLMs or agent systems