Research Engineer (Self-Evolving LLM Systems for Operations Research)
IRG_M3S_T5_2025_015
Project Overview
This role contributes to Towards Self-Evolving LLM Systems for Optimization Problem Formulation under M3S. We build LLMs that evolve both optimization correctness and efficiency by curating an experience library, applying classic OR techniques (e.g., branch-and-cut, column/row generation) and systematically matching the right method to the right problem structure, and supporting human–AI collaboration for large-scale OR problems and real-world tasks (smart cities, transportation, resource allocation).
Key Responsibilities
- Conduct research and develop systems-level prototypes that focus on self-evolving LLM systems for OR, including experience library curation, probabilistic reasoning, solver-confidence integration, and efficiency-aware formulation.
- Publish research results in top-tier conferences and journals, with focus on venues associated with OR, AI and transportation science.
- Collaborate with OR experts and local partners in Singapore to address applied operations scenarios and pilot implementations.
- Assist in grant writing, project management, and other administrative duties related to research activities.
Requirements
- Minimum Bachelor's Degree in Computer Science, Operations Research, Data Science, Transportation Engineering, or related fields.
- Hands-on with LLM agent systems, training/fine-tuning and data science workflows (data preparation/ETL, feature engineering, experiment design and metrics).
- Strong background in OR formulations (e.g., integer programming, submodular optimization) and integration with OR solvers (e.g., Gurobi/CPLEX) are essential.
- Project-level experience with building datasets/benchmarks and efficiency evaluation is highly desired.
- Strong publication record in top-tier AI/ML/OR conferences and journals.
- DESIRABLE: Experience leading human-in-the-loop, systems-level studies using Singapore data and piloting with industrial sector, from dataset/benchmark design to field validation.
- Excellent written/oral communication and cross-functional collaboration skills.
Interested applicants are invited to send in their full CV/resume, cover letter and list of three references (to include reference names and contact information). We regret that only shortlisted candidates will be notified.