- Hong Kong Hong Kong
Anlyvis Limited Hiring! Full Time Vision AI Software Engineer in Hong Kong - Ricebowl
Share
Save
Working Location
Job Description
Responsibilities
Anlyvis develops Edge AIoT and AI-powered video analysis solutions that help cities and enterprises improve safety, security and operations. This role builds the robust perception and temporal-understanding layer behind intelligent video analytics.
Key Responsibilities
Develop real-time detection, tracking, pose, segmentation and action-recognition models.
Build temporal methods that recognise events across consecutive video frames.
Combine tracking, pose, spatial relationships and VLM reasoning for behaviour analysis.
Improve small-object and long-distance performance in demanding camera conditions.
Design data collection, annotation, benchmarking and systematic error analysis.
Integrate CV models with VLMs and optimise pipelines for real-time inference.
Requirements
Degree in computer science, AI, electronic engineering or a related discipline.
Strong computer vision, deep learning, Python and PyTorch fundamentals.
Experience with YOLO, DETR, SAM, ByteTrack, DeepSORT, pose models or similar tools.
Practical experience processing live or recorded video streams.
Understanding of precision, recall, F1, mAP and experiment-driven improvement.
Preferred Qualifications
DeepStream, TensorRT or NVIDIA Jetson deployment experience.
Experience in surveillance, industrial safety, smart-city or retail applications.
Knowledge of VLMs and multimodal scene-understanding methods.
Application Areas
People counting, crowd flow, queues and restricted-area monitoring.
Object, inventory, vehicle, parking and traffic-event analytics.
Gesture, action and unusual-behaviour recognition.
Fire, smoke, fall and industrial-safety monitoring.
How to Apply
Submit your CV, relevant publications or project portfolio, and a short note explaining your interest in Anlyvis and this role.
Full-time
Key Responsibilities
Develop real-time detection, tracking, pose, segmentation and action-recognition models.
Build temporal methods that recognise events across consecutive video frames.
Combine tracking, pose, spatial relationships and VLM reasoning for behaviour analysis.
Improve small-object and long-distance performance in demanding camera conditions.
Design data collection, annotation, benchmarking and systematic error analysis.
Integrate CV models with VLMs and optimise pipelines for real-time inference.
Requirements
Degree in computer science, AI, electronic engineering or a related discipline.
Strong computer vision, deep learning, Python and PyTorch fundamentals.
Experience with YOLO, DETR, SAM, ByteTrack, DeepSORT, pose models or similar tools.
Practical experience processing live or recorded video streams.
Understanding of precision, recall, F1, mAP and experiment-driven improvement.
Preferred Qualifications
DeepStream, TensorRT or NVIDIA Jetson deployment experience.
Experience in surveillance, industrial safety, smart-city or retail applications.
Knowledge of VLMs and multimodal scene-understanding methods.
Application Areas
People counting, crowd flow, queues and restricted-area monitoring.
Object, inventory, vehicle, parking and traffic-event analytics.
Gesture, action and unusual-behaviour recognition.
Fire, smoke, fall and industrial-safety monitoring.
How to Apply
Submit your CV, relevant publications or project portfolio, and a short note explaining your interest in Anlyvis and this role.
Full-time
Important Information
Never provide your bank or credit card details when applying for jobs. Do not transfer any money or complete unrelated online surveys. If you see something suspicious, Report this Job ad.
Learn More