Senior LLM Application Engineer_台達研究院(台北)

We are looking for an experienced LLM Application Engineer to design, build, and operate production-grade AI applications based on Retrieval-Augmented Generation (RAG) and Agentic architectures.

In this role, you will work on real-world enterprise use cases, integrating LLMs with internal systems, APIs, workflows, and automation platforms. You will be responsible not only for model interaction, but also for system reliability, governance, scalability, and security.

This is a hands-on engineering role with strong ownership over architecture and delivery.

Responsibilities:
1. LLM & Agent Application Development

• Design and implement RAG pipelines, including document ingestion, embedding strategies, vector storage, retrieval optimization, and prompt orchestration.
• Build agent-based systems capable of tool calling, multi-step reasoning, and workflow orchestration.
• Develop conversational and task-oriented AI agents that interact with enterprise systems, databases, and APIs.

2. System Integration & Automation

• Integrate LLM applications with existing enterprise platforms (ERP, CRM, BI, internal services).
• Implement agentic workflows combined with automation tools (e.g., RPA, workflow engines, or low-code platforms).
• Design API-based integrations for third-party services and internal microservices.

3. Architecture & Production Readiness

• Design scalable, maintainable architectures for LLM applications running in production.
• Implement logging, monitoring, evaluation, and error-handling for AI agents.
• Ensure solutions meet security, governance, and compliance requirements (e.g., access control, data isolation, auditability).

4. DevOps & Delivery

• Containerize and deploy LLM services using Docker and CI/CD pipelines.
• Collaborate with product, operations, and business stakeholders to translate requirements into reliable AI systems.
• Take technical ownership from prototype to production rollout.