Senior AI Engineer (Computer Vision & Intelligent Systems)

EduRankAI; Kolkata District; Presencial; INR 4.000.000 - 7.000.000; Jornada completa

Descripción de la vacante

EduRankAI in India is seeking a senior engineer to own the end-to-end path from camera input to real-time decision in production computer-vision systems, including detection, tracking, and 3D vision.

You will design real-time pipelines, optimize models for latency budgets, integrate with proctoring and analytics platforms, and mentor mid-level engineers to raise the technical bar.

Formación

• Shipped production computer-vision systems, not notebooks.
• Experience with real-time vision pipelines and latency budgets.
• Familiar with multi-camera and 3D vision concepts.

Responsabilidades

• Design and ship real-time computer-vision pipelines: object detection, tracking, pose estimation, OCR, and video analytics.
• Own model selection, training, optimisation, and deployment (quantisation, GPU inference, edge targets).
• Build multi-camera and 3D-vision systems: calibration, reconstruction, trajectory estimation, sensor fusion.
• Integrate vision systems with our proctoring, assessment, and sports platforms with strict latency budgets.
• Establish evaluation harnesses: dataset curation, ground truth, drift monitoring, and failure analysis.
• Mentor mid-level engineers and set the technical bar for applied vision work.

Conocimientos

Python

C++

PyTorch

OpenCV

TensorRT

Object detection

Pose estimation

Multi-camera calibration

3D reconstruction

Real-time systems

GPU programming

Docker

Kubernetes

Linux

CUDA

Herramientas

Docker

Kubernetes

PyTorch

OpenCV

TensorRT

CUDA

Descripción del empleo

Build production computer-vision systems — detection, pose, video analytics, 3D vision — that run in real time.

Much of the most valuable AI work is visual: proctoring integrity, sports analytics, document understanding, industrial inspection, immersive learning. We are hiring a senior engineer who has shipped production computer-vision systems — not notebooks — and who is comfortable owning the full path from camera to decision. Strong engineers in this specialisation often skip listings that only say "LLM Engineer"; this role exists so they do not have to.

Responsibilities

• Design and ship real-time computer-vision pipelines: object detection, tracking, pose estimation, OCR, and video analytics.
• Own model selection, training, optimisation, and deployment (quantisation, GPU inference, edge targets).
• Build multi-camera and 3D-vision systems: calibration, reconstruction, trajectory estimation, sensor fusion.
• Integrate vision systems with our proctoring, assessment, and sports platforms with strict latency budgets.
• Establish evaluation harnesses: dataset curation, ground truth, drift monitoring, and failure analysis.
• Mentor mid-level engineers and set the technical bar for applied vision work.

Skills

• Python
• C++
• PyTorch
• OpenCV
• TensorRT or equivalent inference optimisation
• Object detection (YOLO-family, transformer detectors)
• Pose estimation
• Multi-camera calibration & 3D reconstruction
• Real-time systems
• Docker & Kubernetes
• Linux
• GPU programming (CUDA a plus)