Lead Research Engineer I

A*STAR - Agency for Science, Technology and Research; Singapore; On-site; SGD 120,000 - 180,000; Full time

Job summary

A*STAR's ARTC Digital Supply Chain Group seeks an engineer with expertise in agentic AI, LLMs and demand planning analytics to help design next‑generation AI‑powered decision‑support for supply chains across industries.

You will build end‑to‑end solutions connecting AI agents with forecasting models, databases, APIs, and enterprise data, and collaborate with domain experts to deliver user‑centric dashboards and validated, explainable recommendations.

Qualifications

• Bachelor’s or Master’s degree in a related field such as CS, data science, AI, industrial engineering, OR, or SCM.
• Strong proficiency in Python and data analytics/AI libraries.
• Hands-on experience with LLM architectures, agentic workflows, backend/frontend integration.
• Experience with LangChain, RAG, tool calling, and context management.
• Ability to connect AI agents with enterprise data, APIs, and business rules.

Responsibilities

• Design and develop agentic AI solutions for demand planning and supply-chains.
• Develop LLM-powered agents and workflows that reason over data and generate actionable recommendations.
• Apply LLM/agentic AI technologies including tool calling, RAG, and human-in-the-loop controls.
• Integrate LLM agents with forecasting models, enterprise data, APIs, and business rules.

Skills

Python

LLMs

Agentic AI

Demand planning analytics

Frontend development

Backend development

REST APIs

Time-series forecasting

Data analytics

Communication skills

Education

Bachelor’s/Master’s in CS/Data Science/AI/Industrial Engineering/Operations Research/Supply Chain Management

Tools

LangChain

LangGraph

LlamaIndex

Semantic Kernel

FastAPI

React

PyTorch

Job description

Job Description

The Digital Supply Chain Group at the Digital Manufacturing Division, ARTC, is seeking an Engineer with expertise in Agentic AI, Large Language Models (LLMs), demand planning analytics, and AI application development. This role will contribute to the development of next-generation agentic demand planning and supply chain decision-support solutions.

The successful candidate will design and develop LLM-powered agents and agentic workflows that integrate demand forecasting, demand sensing, supply-demand analysis, material shortage diagnostics, and data-driven recommendations. The candidate should have hands‑on experience with recent LLM architectures, agentic workflows, backend development, and frontend integration, with the ability to connect AI agents with analytical models, enterprise data, APIs, and business rules.

The candidate will work closely with supply chain domain experts, AI scientists, and engineers to develop end‑to‑end, user‑centric AI solutions and apply emerging LLM and agentic AI technologies to real‑world demand planning and supply chain challenges across multiple industries.

Job Responsibilities

• Design and develop agentic AI solutions for demand planning, integrating demand forecasting, demand sensing, supply‑demand analysis, and planning recommendations.
• Develop LLM-powered agents and agentic workflows that reason over supply chain data, invoke analytical tools and models, and generate actionable recommendations.
• Apply recent LLM and agentic AI technologies, including tool/function calling, RAG, structured outputs, context management, agent orchestration, and human‑in‑the‑loop workflows.
• Integrate LLM agents with forecasting models, optimization algorithms, enterprise data, APIs, and business rules to support end‑to‑end planning processes.

Demand Planning Analytics

• Develop and enhance demand forecasting and predictive analytics models, including data preparation, feature engineering, model evaluation, and forecast accuracy analysis.
• Analyze demand changes, forecast deviations, supply‑demand imbalances, and material shortage risks and assess their downstream impact.
• Develop data‑driven purchase and mitigation recommendations based on demand, inventory, supply availability, lead times, and operational constraints.

AI Application Development

• Develop backend services and APIs integrating LLMs, AI agents, analytical models, databases, and enterprise systems.
• Develop and integrate interactive frontends and dashboards that enable planners to review forecasts, investigate demand changes, interact with AI agents, and evaluate recommended actions.
• Build end‑to‑end prototypes connecting frontend applications, backend services, agentic workflows, analytical models, and enterprise data.
• Implement appropriate validation, monitoring, and human‑in‑the‑loop controls to improve the reliability and explainability of AI-generated recommendations.

Research & Collaboration

• Evaluate and apply emerging developments in LLMs, agentic AI, multi‑agent systems, and AI application architectures to supply chain use cases.
• Work closely with supply chain experts, AI scientists, data engineers, and software developers to translate business challenges into practical AI‑driven solutions.
• Contribute to research and industry projects in agentic demand planning and AI‑driven supply chain decision support.

Job Requirements

• Bachelor’s/Master’s degree in Computer Science, Data Science, Artificial Intelligence, Industrial Engineering, Operations Research, Supply Chain Management, or a related field
• Strong proficiency in Python and relevant data analytics/AI libraries such as Pandas, NumPy, Scikit‑learn, PyTorch, or equivalent frameworks
• Hands‑on experience developing applications using recent Large Language Model (LLM) architectures and technologies, including prompt engineering, structured outputs, tool/function calling, retrieval‑augmented generation (RAG), and context management
• Practical experience designing and implementing agentic AI or multi‑agent workflows, including agent orchestration, tool integration, planning/reasoning, workflow management, and human‑in‑the‑loop mechanisms
• Familiarity with modern LLM and agentic application frameworks such as LangChain, LangGraph, LlamaIndex, Semantic Kernel, AutoGen, or equivalent technologies
• Experience integrating LLMs and AI agents with analytical models, databases, APIs, enterprise systems, and external tools to build end‑to‑end AI applications
• Strong backend development experience, including development of REST APIs, microservices, data services, and application/business logic using frameworks such as FastAPI, Flask, or equivalent technologies
• Experience in frontend development and integration for AI applications using technologies such as React, Streamlit, Dash, Plotly, PyQt, or equivalent frameworks
• Ability to develop end‑to‑end prototypes, connecting frontend interfaces, backend services, LLM/agentic workflows, analytical models, databases, and enterprise systems
• Relevant experience in machine learning and predictive analytics, preferably involving time‑series forecasting, demand forecasting, optimization, or other supply chain applications
• Understanding of demand planning and supply chain analytics, including demand signals, forecast accuracy, inventory, material requirements, supply‑demand balancing, and related supply chain KPIs
• Familiarity with enterprise and supply chain systems such as SAP, ERP, MRP, MES, or related planning and transactional data sources is highly preferred
• Experience implementing appropriate LLM guardrails, output validation, observability, evaluation, and monitoring to improve the reliability and traceability of agentic AI applications
• Strong analytical and problem‑solving skills, with the ability to translate complex business and planning problems into AI‑enabled workflows and practical decision‑support solutions
• Excellent interpersonal and communication skills with the ability to work effectively with supply chain domain experts, AI researchers, data engineers, software developers, and industry stakeholders
• Bonus: Familiarity with LLM evaluation, reasoning models, MCP/tool integration, vector databases, knowledge graphs, agent memory, workflow automation, explainable AI, and cloud/container deployment is advantageous