Lead Computer Vision Engineer - Real-Time Multi-Camera Analytics

Jobline Resources Pte Ltd; Singapore; On-site; SGD 120,000 - 170,000; Full time

Job summary

Jobline Resources Pte Ltd is seeking an experienced software engineer to architect and optimize large-scale video analytics pipelines for live camera feeds. You will implement cross-camera tracking, configure real-time streaming, and build integration points with VMS and access control systems.

The role emphasizes performance tuning, embedding models with NVIDIA DeepStream, and delivering sub-second latency from sensor to dashboard across enterprise environments in Singapore.

Qualifications

• 5+ years of software engineering experience, with 3+ years in large-scale video analytics pipelines.
• Proficiency in Python and performance-critical C++.
• Deep expertise in PyTorch/TensorFlow, DeepStream, GStreamer, Docker, and distributed queues (Kafka).
• Strong grasp of real-time streaming protocols (RTSP, WebRTC, SRT, HLS) and secure transport configurations.

Responsibilities

• Architect C2-scale analytics pipelines using Python/C++ across hundreds of live camera feeds.
• Translate model outputs into lightweight structured metadata payloads (JSON/Protobuf).
• Stream alerts to C2 brokers via Kafka, MQTT, or gRPC.
• Develop multi-camera tracking and fusion algorithms for cross-camera re-identification.
• Integrate with VMS, NVRs, and access control layers via plugins and middleware.
• Profile and optimize inference pipelines using TensorRT/ONNX Runtime to meet latency targets.

Skills

Python

C++

Video analytics

Distributed queues

Realtime streaming

Tools

NVIDIA DeepStream SDK

GStreamer

PyTorch

TensorFlow

OpenCV

Docker

Kafka

Job description

Jobline Resources Pte Ltd is seeking an experienced software engineer to architect and optimize large-scale video analytics pipelines for live camera feeds. You will implement cross-camera tracking, configure real-time streaming, and build integration points with VMS and access control systems.

The role emphasizes performance tuning, embedding models with NVIDIA DeepStream, and delivering sub-second latency from sensor to dashboard across enterprise environments in Singapore.