Senior Applied AI Engineer, Agentic Systems

Agilent helps laboratories around the world advance scientific discovery, diagnostics, and applied market solutions through instruments, software, consumables, services, and deep domain expertise.

About the role

As a Senior Applied AI Engineer, Agentic Systems, you will be an AI engineering SME within a cross-functional AI pod, working alongside data engineers, domain experts, business stakeholders, and platform teams. Your role is to design, build, deploy, and govern AI solutions that solve real scientific, service, commercial, and operational challenges across the enterprise. 

You will work directly with the end users to understand their workflows, translate business needs into production-grade AI solutions, and ensure those solutions are reliable, observable, reusable, and responsible.

This role goes beyond prototyping. You will build AI agents, RAG applications, orchestration patterns, evaluation frameworks, and reusable skills that become part of Agilent’s broader AI ecosystem. 

You do not need to know Agilent’s internal AI platform terminology on day one. We are looking for someone with strong software engineering foundations, experience delivering AI applications into production, and a passion for building practical solutions that create measurable business value. 

What you will do:

• Define the architecture and build approach for the pod's use case, selecting appropriate agents, RAG, retrieval, and orchestration patterns; designs are reviewed with the Head of Agentic AI Platform Engineering to ensure alignment with reference patterns and platform standards. 

• Build reusable AI skills and integrations using shared platform services and established integration patterns rather than one-off solutions, ensuring components are independently testable and designed for reuse, with appropriate identity, access, and governance controls.

• Embed evaluation, testing, and observability into delivery from the first sprint, defining benchmark sets with domain experts and establishing regression testing and production monitoring before launch. 

• Implement AI solutions in accordance with security, compliance, and governance requirements, including risk-based controls and human-in-the-loop approval points for regulated or GxP-relevant use cases. 

• Contribute reusable skills, orchestration patterns, and evaluation assets back to the broader AI platform, documenting them for future teams and designing with the second consumer in mind. 

• Partner with and upskill domain SMEs and business practitioners within the pod, helping them sustain AI solutions and providing feedback back to improve platform capabilities over time.

What success looks like in year one

• The pod's agentic system is live in production within the domain workflow, with documented governance, identity controls, and human-in-the-loop processes where required. 

• Evaluation coverage is established for critical AI behaviors, with testing, monitoring, and observability embedded into the delivery lifecycle. 

• Multiple reusable skills, orchestration patterns, or evaluation assets created by the pod have been adopted, reused, or positioned for reuse across additional use cases.

• A rotating SME or business practitioner from the pod has increased their AI capability through active contribution to the solution and ongoing adoption of AI-enabled ways of working.