Research Engineer, Advanced Sensing & Smart Inspection, ARTC

Responsibilities

Develop AI/ML models and agent modules; Train, validate, and evaluate AI models using sensor / inspection data, simulations outputs, and experiments.

Translate partner and customer needs into end-to-end inspection solutions linking data acquisition, hardware, inference, and decision-making.

Support experimental setup, calibration, data acquisition, and hardware troubleshooting.

Support hardware-software integration with cameras, illumination, sensors, robotic or motion stages, programmable controllers, and edge-computing devices.

Assist in setup, calibration, data acquisition, and hardware troubleshooting to ensure reliable data for AI model development and validation.

Prototype, test, and iterate AI solutions, with a focus on progressing successful concepts toward deployment.

Collaborate with teams across simulation, optimisation, knowledge storage, hardware integration, and validation.

Document research findings, model performance, validation results, and system architectures.

Keep abreast of advances in agentic AI, machine learning, computer vision, and automation to bring relevant new ideas to the project

Bachelor's or Master's degree in Computer Science, Engineering, AI, Robotics, Mechanical Engineering or a related field.

Strong AI/ML foundation, including model training, evaluation, and practical implementation.

Proficient in Python, with experience applying AI/ML in coursework, projects, internships, or research.

Solid software development fundamentals, including version control and maintainable code practices.

Basic understanding of hardware-software integration, such as sensors, cameras, edge devices, or experimental setups.

Self-motivated, analytical, and able to collaborate effectively with cross-functional teams.

Preferred Knowledge and Skills

Hands-on experience with PyTorch, TensorFlow, or similar AI/ML frameworks.

Experience with computer vision libraries and techniques (e.g., OpenCV, object detection, segmentation, defect classification).

Familiarity with, or strong interest in learning, agentic AI patterns, LLM-based frameworks such as LangChain or LangGraph, and multi-agent orchestration.

Exposure to knowledge representation, vector databases, or retrieval-augmented generation (RAG).

Experience taking prototypes to production, including automation, deployment, integration, or MLOps practices.

Background in signal processing, robotics/automation, or manufacturing quality inspection. Domain knowledge is beneficial but can be developed on the job.

Experience with inspection hardware such as cameras, lighting, sensors, motion stages, robotics, PLCs, microcontrollers, or edge AI devices, including setup, integration, calibration, or troubleshooting.

The above eligibility criteria are not exhaustive. A*STAR may include additional selection criteria based on its prevailing recruitment policies. These policies may be amended from time to time without notice. We regret that only shortlisted candidates will be notified.