Permanent Perception Engineer (Computer Vision & Multimodal AI) at Ang Mo Kio , Up to $5,000 (Office hours)

• Permanent
• Up to $5,000
• Mon-Fri: 8.30am-5pm
• Ang Mo Kio

Role Summary Responsible for developing, evaluating and maturing applied AI, computer vision and multimodal perception capabilities for real-world surveillance applications. The engineer will work closely with the robotics and software teams to deploy validated perception functions into AST systems. The role focuses on perception models, data/evaluation and multimodal AI , and is an applied engineering position rather than a pure AI research or data-science role. Key Responsibilities: Develop and enhance computer vision capabilities including object detection, classification, multi-object tracking and false-alert reduction . Work with EO/IR and thermal imagery , and where applicable combine perception outputs with AIS, radar tracks and other system metadata. Build representative datasets, annotation practices, evaluation sets and repeatable performance benchmarks suitable for field conditions. Analyse false positives, false negatives, dataset shifts and operational failure modes , and translate findings into measurable improvements. Develop and evaluate multimodal AI, behaviour/event analytics and operator-assist functions where they provide operational value. Optimise and package AI models for practical edge deployment using ONNX, TensorRT/CUDA or equivalent technologies. Collaborate closely with the Robotics Software Engineer on model deployment and system integration. Support controlled experimentation with emerging technologies including vision-language models, synthetic data and active learning . Support integration testing and field trials to validate perception capabilities under operational conditions.

Period: Permanent

Location: Ang Mo Kio

Working Hours: Monday – Friday, 8.30am – 5pm

Salary: Up to $5,000

Job Requirements: Degree in Computer Science, Artificial Intelligence, Computer/Electrical Engineering, Robotics or related discipline.

Strong foundation in Python with practical machine-learning and computer-vision development experience.

C++ capability would be advantageous.

Hands-on experience in one or more of the following:

• Object detection
• Multi-object tracking
• Image/video analytics
• Thermal imaging
• Multimodal AI
• Sensor fusion

Comfortable working with datasets, model evaluation, error analysis and quantitative performance metrics . Experience with PyTorch or equivalent, OpenCV, ONNX, TensorRT/CUDA, NVIDIA Jetson or edge deployment would be advantageous. Candidates with strong relevant project, internship or practical engineering experience may also be considered. Applied hands-on capability is more important than years of experience alone.  Fresh grads with relevant internship and project experience are welcome to apply

Other Remarks: Work closely with internal systems, software platform, sensor-data, robotics software and verification teams . Take primary ownership of applied AI/perception model development and multimodal analytics . focusing on AI models, perception performance, datasets and analytics .

Interested applicants, kindly email your detailed resume ( MS Word format is preferred): azmin@successhrc.com.sg (Reg No: R25129300) Please ensure that applications sent through email are no bigger than 1Mb. We thank all applicants for your interest but regret to inform that only shortlisted candidates would be notified. Success Resource Centre Pte Ltd (EA License Number: 04C3201) 160 Robinson Road, #13-07/08/09 SBF Center, Singapore 06891 T: 6337 3183 | W: www.successhrc.com.sg 

By submitting your application for this position, you consent to the collection, use, and disclosure of your personal data by SHRC and its recruitment partners, including Employment and Employability Institute Pte Ltd, for the purpose of i) employment facilitation and ii) data and statistical analysis. You may find out more information on the PDPA policies at https://successhrc.com.sg/data-protection-notice/ and https://www.e2i.com.sg/pdpa/.