Computer Vision Engineer

Motivity Labs; Hyderabad; On-site; INR 1,800,000 - 2,600,000; Full time

Job summary

Motivity Labs is seeking a Computer Vision Engineer specializing in real-time video analytics and edge AI systems. You will design, optimise, and deploy high-performance vision pipelines across edge devices and cloud environments.

The role focuses on model building, training, optimisation, and scalable AI deployment for tasks such as object detection, pose estimation, face recognition, and behavior analytics. Strong experience in production ML systems is required.

Qualifications

• Hands-on experience in real-time computer vision systems and edge AI deployment.
• Strong proficiency with Python and C++, plus familiarity with Java is expected.
• Deep expertise in OpenCV, PyTorch or TensorFlow, TensorRT, OpenVINO, and DeepStream.

Responsibilities

• Design, train, and optimise deep learning models (YOLO, CNNs) for vision tasks.
• Develop real-time inference pipelines and edge deployments on Jetson/ARM/NPUs.
• Benchmark FPS, latency, throughput, memory, and accuracy in production setups.
• Collaborate with cross-functional teams to integrate AI into production systems.

Skills

Computer Vision

Python

C++

OpenCV

PyTorch

TensorFlow

TensorRT

OpenVINO

DeepStream

Edge AI

RTSP

CUDA

Docker

Job description

About the Role :

We are looking for a Computer Vision Engineer specialising in real-time video analytics and edge AI systems. The role involves designing, optimising, and deploying high-performance vision pipelines across edge devices (Jetson, ARM, CPU/NPUs) and cloud environments.

This role requires strong expertise in real-time vision systems, model building, model training, model optimisation, and scalable AI deployment. You will work on real-world problems involving object detection, pose estimation, face recognition, behavior analytics, and large-scale inference optimisation.

What You'll Do:

Model Development & Training:

Design, train, and optimise deep learning models (YOLO, CNNs) for computer vision tasks

Preprocess, augment, and curate large-scale image and video datasets

Inference & Optimisation:

Build end-to-end inference pipelines using TensorRT, OpenVINO, and CUDA

Optimise models from FP32 to FP16/INT8 for edge and production environments

Benchmark system performance across FPS, latency, throughput, memory, and accuracy

Edge & Real-Time Systems:

Develop low-latency inference pipelines for edge devices (Jetson, ARM CPUs, NPUs)

Process real-time RTSP video streams at production scale

Collaboration & Growth:

Work with cross-functional teams to integrate AI into production systems

Debug and resolve issues related to training pipelines, environments, and runtime performance

Stay current with advancements in computer vision and edge AI.

Required Skills:

Core Skills:

46 years of hands-on Computer Vision experience

Strong proficiency in Python and C++; familiarity with Java

Deep expertise in OpenCV, PyTorch/TensorFlow, TensorRT, OpenVINO, and DeepStream

Solid understanding of image processing algorithms and model evaluation metrics

Production Experience:

Real-time RTSP stream processing

Deploying and maintaining production-grade ML systems

Edge AI deployment with hardware-aware optimisation

Nice to Have:

Docker and containerised deployment workflows

Cloud platforms (AWS, Azure, or GCP)

Exposure to surveillance, VMS, or CCTV analytics domains