Real-Time Computer Vision Engineer: Vision Models & Streams

Cognizant; Ciudad de Mendoza; On-site; ARS 1,800,000 - 3,200,000; Full time

Job summary

Cognizant in Mendoza, Argentina seeks a hands-on AI Computer Vision Engineer to design and deploy production-grade vision models for real-time attribute extraction from video streams.

You will build low-latency pipelines, optimize inference on cloud (Kubernetes/GCP) and edge devices, work with LPR/OCR and state-of-the-art detectors, and contribute to dataset curation and active learning loops with PyTorch and OpenCV. 3+ years of practical CV experience required.

Qualifications

• 3+ years of hands-on experience designing and deploying computer vision models in production.
• Deep expertise with modern object detectors (YOLO, DETR, Faster R-CNN, ViT).
• Strong programming in Python and DL frameworks (PyTorch, OpenCV, TensorFlow).
• Proven experience with video stream ingestion, frame processing and low-latency inference.
• Experience with LPR, OCR, or attribute inference systems.
• Familiarity with quantization, ONNX export, and TensorRT compilation.

Responsibilities

• Model Architecture & Development: design, train, and fine-tune vision models for object detection and attribute extraction.
• Stream Processing & Pipeline Engineering: build low-latency ingestion and processing pipelines for video streams.
• Plate Recognition, OCR, and multi-object tracking.
• Inference Optimization: optimize performance on cloud (Kubernetes/GCP) and edge devices using TensorRT/ONNX/OpenVINO.
• Dataset & Active Learning: curate and augment data, setup automated retraining loops and evaluation.

Skills

Python

Deep learning

YOLO/DETR

Video processing

LPR/OCR

TensorRT

ONNX export

Tools

PyTorch

OpenCV

TensorFlow

ONNX

TensorRT

Job description

Cognizant in Mendoza, Argentina seeks a hands-on AI Computer Vision Engineer to design and deploy production-grade vision models for real-time attribute extraction from video streams.

You will build low-latency pipelines, optimize inference on cloud (Kubernetes/GCP) and edge devices, work with LPR/OCR and state-of-the-art detectors, and contribute to dataset curation and active learning loops with PyTorch and OpenCV. 3+ years of practical CV experience required.