Research Fellow — Agentic AI & Multi-Agent RAG Systems

Nanyang Technological University Singapore; Singapore; On-site; SGD 70,000 - 110,000; Full time

Job summary

NTU's Advanced Technology and Medical Research Institute (ATMRI) invites applications for a Research Fellow to lead the Digital ATCO Assistant project. You will drive knowledge, retrieval, reasoning, and agent orchestration using LLMs, Agentic AI, and multi-agent architectures.

You will implement tool calls and end-to-end RAG pipelines across aviation domains and operate within APIs/microservices. The role emphasizes collaboration with domain experts, controllers, and peers to shape research

Qualifications

• PhD or Master’s with substantial applied research in AI/ML/NLP.
• 2-4 years hands-on with LLMs and Agentic AI.
• Experience with RAG, vector databases, and knowledge graphs.
• Strong Python, PyTorch, and API/microservices skills.

Responsibilities

• Lead design and evaluation of Agentic AI and multi-agent architectures for the Digital ATCO Assistant.
• Develop tool calling to invoke tools, APIs, and knowledge services with validation and error handling.
• Build end-to-end RAG pipelines across AIPs, NOTAMs, and procedures using vector databases and knowledge graphs.
• Design benchmarks for task performance, coordination, and factual grounding.
• Collaborate with engineers and domain experts to expose capabilities via APIs and microservices.
• Translate operational use cases into research requirements, datasets, and test scenarios.
• Contribute to reports, proposals, publications, and collaborations with partners.

Skills

LLM

Agentic AI

Multi-agent architectures

Tool calling

Prompt engineering

Education

PhD in AI/CS/ML/NLP

Master’s with applied research

Tools

Python

PyTorch

LLM frameworks

APIs

Microservices

Vector databases

RAG

Job description

NTU's Advanced Technology and Medical Research Institute (ATMRI) invites applications for a Research Fellow to lead the Digital ATCO Assistant project. You will drive knowledge, retrieval, reasoning, and agent orchestration using LLMs, Agentic AI, and multi-agent architectures.

You will implement tool calls and end-to-end RAG pipelines across aviation domains and operate within APIs/microservices. The role emphasizes collaboration with domain experts, controllers, and peers to shape research