Sr. AI Agent / Prompt Engineer
Job Title\: Sr. AI Agent / Prompt Engineer
Hyundai Capital America (HCA) helps people move forward. Through Hyundai Motor Finance, Genesis Finance, and Kia Finance, we deliver innovative financing, leasing, and insurance solutions to more than 3 million customers and businesses nationwide.
We’re a company driven by growth, innovation, and people. At HCA, you’ll find opportunities to build new skills, expand your career, and make a real impact—while working in a diverse, inclusive, and values‑driven environment. We’re proud to support our communities through volunteerism, philanthropy, and engaged Employee Resource Groups.
If you’re looking for a fast‑paced, collaborative workplace where your ideas matter, join us as we lead the future of financing freedom of movement. Apply today.
WORK MODEL
#LI-OnSite (4 days a week)
WHAT YOU WILL DO
The Sr. AI Agent / Prompt Engineer designs, builds, and hardens production grade AI agents that use large language models to reason, invoke tools, retrieve enterprise knowledge, and recover from errors. This role goes beyond prompt writing and owns agent behavior end to end, including prompt systems, tool orchestration, guardrails, evaluation, and reliability in real business workflows. This role will operate independently, partner closely with architecture, product, governance, and UX, and actively drive AI initiatives from prototype through production.
HOW YOU WILL MAKE AN IMPACT
1. Agent and Prompt Engineering
• Design, build and deploy multi‑step prompt chains for complex of reasoning, planning, tool integration, and error recovery ensuring accuracy, reliability and contextual awareness.
• Author and maintain robust prompt libraries (system prompts, task prompts, tool prompts) with clear structure, constraints, and fallbacks.
• Develop agent workflows using function calling, graph‑based orchestration, or similar agent frameworks.
• Continuously iterate prompts and agent logic to improve accuracy, consistency, and grounded response.
2. Tool Use and Integration
• Integrate agents with enterprise tools, APIs, and data sources (e.g. retrieval systems, workflow engines, internal services).
• Define safe tool use patterns, permissions, and auditing strategies in partnership with governance and security teams.
• Partner with data and platform engineers to ensure reliable RAG and agent execution paths.
3. Evaluation, Quality, and Reliability
• Define and run offline and online evaluations for agent behavior (task success, tool success, error recovery).
• Diagnose failure modes such as hallucinations, prompt injection, and tool misuse.
• Partner with QA and Governance to harden agents prior to production release.
4. Cross-Functional Collaboration
• Partner with business teams and product managers to translate business use cases into agent behaviors and integrate AI agents into production systems.
• Ensure optimized design for interaction flows, transparency, and user trust
• Stay current of emerging AI research trends and recommend practical improvements to the enterprise AI stack.