Research Engineer/Assistant (Agentic AI & Urban Intelligence)-Cities Foresight Lab(CFL), NUS Cities

Interested applicants are invited to apply directly at the NUS Career Portal. Please note your application will only be processed if you apply via NUS Career Portal.

NUS Career Portal link: https://careers.nus.edu.sg/job/Research-EngineerAssistant-%28Agentic-AI-&-Urban-Intelligence%29-Cities-Foresight-Lab%28CFL%29%2CNUS-Cities/32682-en_GB/

We regret that only shortlisted candidates will be notified.

Job Description

NUS Cities Foresight Lab (CFL) is seeking a Research Engineer/Assistant to develop an Agentic Orchestration Framework for Urban Intelligence. The project aims to demonstrate how agentic systems can enhance efficiency and consistency in policy and planning workflows by synthesizing unstructured datasets (e.g., text, social media, regulatory frameworks) with structured spatiotemporal information to generate responsive, data-driven recommendations.

You will be responsible for implementing a scalable Agentic Framework with Model Context Protocol (MCP) integration to connect diverse data sources with expert agent models. The role involves developing an orchestration layer capable of transforming complex, multi-scale technical inputs into actionable, reasoned insights through a natural-language interface. You will work alongside urban planning and social science researchers to ensure the system's outputs are interpretable, context-aware, and relevant to real-world policy and planning workflows.

Key Responsibilities
. Agentic Framework Design: Develop and maintain a multi-step reasoning framework (e.g., LangGraph or similar) that can autonomously decompose high-level user objectives into executable tasks.
. MCP Integration: Implement and scale Model Context Protocol (MCP) servers to standardize the interface layer between external data repositories, real-time APIs, and specialized analytical models.
. Urban Intelligence Case Demonstration: Perform data synthesis for case studies, spatiotemporal tool engineering, and expert agent tuning.
. Performance Evaluation: Define and track system performance (e.g., API compatibility across system architecture, step tracing) and user validation metrics (e.g., ground-truthing, benchmarking against manual workflows).
. Documentation & Publication: Contribute to data/method documentation, visualisations, and writing reports/publications.

Qualifications

Qualifications
. Bachelor's or Master's Degree in Computer Science, Artificial Intelligence, Data Science, or a related quantitative field with a strong computational focus.
. Proficiency in Python.
. Hands-on experience with agentic frameworks, specifically in designing multi-step reasoning loops and tool-calling logic, and system architecture, including MCP.
. Experience in data engineering, including the ability to handle both unstructured and structured datasets.
. Familiarity with natural language processing tasks such as sentiment analysis, knowledge bases, and information retrieval.
. Resourceful and critical with good communication skills able to work independently while collaborating effectively with interdisciplinary teams of urban planners and social scientists.

Preferred
. Experience in retrieval-augmented generation (RAG), knowledge graphs, or structured reasoning over heterogeneous data sources.
. Familiarity with cloud deployment environments and modern software development practices.
. Familiarity with geospatial data analysis and libraries (e.g., GeoPandas, Shapely, or equivalent).
. Familiarity with explainability and trust frameworks for AI systems, particularly in public sector or governance contexts.
. Interest in urban, social, or policy applications of data science and AI.

Application Procedure
Interested applicants should submit a dossier consisting of the following to the NUS Career Portal:
. a cover letter (maximum 3 pages)
. up-to-date CV
. a statement describing their research trajectory, interests and career ambitions
. contact details for three referees. Only short-listed applicants will be invited to submit reference letters.

The anticipated start date for the position is 1 June 2026. We will begin evaluating candidates immediately, but the position will remain open until a suitable candidate is found. Further enquiries should be sent to [Confidential Information] (please indicate Research Assistant (Agentic AI & Urban Intelligence) Application as the subject heading).