Senior Deep Learning Engineer

Job Description

We are seeking an experienced Deep Learning and Computer Vision Consultant to lead strategic AI initiatives, design scalable model architectures, and build cutting-edge visual intelligence solutions. This role demands technical depth, production-grade engineering skills, and the ability to operate at the intersection of research, enterprise AI, and real-world deployment.

Key Responsibilities

AI Strategy Leadership:
Define and drive the roadmap for Computer Vision and Deep Learning initiatives aligned with organizational and client objectives.

Model Development:
Design, train, and deploy advanced vision and multimodal models using architectures such as Vision Transformers, SAM, and multimodal LLMs.

Video Analytics & Edge AI:
Build real-time analytics solutions using NVIDIA DeepStream, including multi-camera pipelines, person tracking, object detection/segmentation, and smart city–scale video AI workflows.

Integration & Optimization:
Develop high-performance inference pipelines using frameworks like vLLM and optimize models for edge, on-prem, and cloud environments (AWS, GCP, Azure).
Perform model compression, quantization, TensorRT optimization, and GPU utilization tuning.

End-to-End ML Lifecycle:
Manage the full ML lifecycle—from data curation/annotation to training, experimentation, versioning, deployment, and monitoring in production.

Production Deployment:
Architect and deploy production-grade AI systems across on-premise GPU servers, edge devices, and cloud-native stacks using Docker, Triton, TorchServe, or equivalent serving frameworks.

Cross-Functional Collaboration:
Collaborate with product teams, data engineers, DevOps, and business stakeholders to translate requirements into scalable AI capabilities.

Requirements

Experience:
6–8 years of hands-on experience in Deep Learning, Computer Vision, and applied AI systems, including real-time video analytics.

Technical Proficiency:
Strong command of PyTorch, TensorFlow, OpenCV, ONNX, Hugging Face, and GPU-accelerated model development.

Video AI & DeepStream Expertise:

• Hands-on experience with NVIDIA DeepStream, GStreamer pipelines, and GPU-accelerated video inference.
• Experience in person tracking, re-identification (ReID), multi-camera analytics, and smart city video workflows.

Deployment Expertise:
Proven experience deploying models at scale using Docker, Kubernetes, APIs, and GPU-optimized serving frameworks such as Triton or TorchServe.

Model Optimization:
Proficiency in quantization, pruning, TensorRT optimization, batching strategies, and throughput tuning for real-time inference.

On-Prem & Edge Deployment:
Experience deploying on on-premise GPU clusters, edge devices (Jetson series or similar), and integrating with multiple CCTV streams.

MLOps Knowledge:
Familiarity with CI/CD for ML, data pipelines, monitoring, MLFlow/Weights & Biases, and production observability.

Soft Skills:
Strong analytical thinking, communication, problem-solving, and leadership. Passion for innovation and mentoring team members.

Why Join Us

At Clarion Analytics, you’ll work with a forward-thinking AI team building next-generation computer vision and multimodal systems for real-world enterprise and smart city applications.

If you’re passionate about production-grade AI engineering, real-time video analytics, and pushing the boundaries of intelligent vision systems, we would love to collaborate.

Learn more: https://clarion.ai
Join our community: https://www.linkedin.com/company/clarion-analytics-singapore

Job Type: Freelance
Contract length: 12 months

Pay: ₹500.00 - ₹750.00 per hour

Expected hours: 2 – 10 per week

Experience:

• Deep learning: 4 years (Preferred)
• Computer vision: 4 years (Preferred)
• PyTorch: 4 years (Preferred)