Mid-level Computer Vision/Machine Learning (Integração e Deployment)

12 days ago

Recife, BrazilMid Level

Responsibilities

Develop and integrate image and video acquisition, processing, and analysis pipelines.
Integrate computer vision algorithms and machine learning models into industrial inspection solutions.
Develop production image-processing and inference solutions for cloud and edge environments.
Optimize models and pipelines for performance, latency, and efficient computational-resource usage.
Implement inference solutions using ONNX and ONNX Runtime.
Develop APIs and integration interfaces between solution components.
Implement containers and reproducible execution environments for algorithms and models.
Contribute to CI/CD pipelines and automate build, testing, and deployment processes.
Implement logging, monitoring, and diagnostic mechanisms.
Collaborate with data scientists, computer vision specialists, software engineers, and other project professionals.
Support model and algorithm evolution using information gathered during solution execution.
Document architecture, components, interfaces, deployment processes, and technical decisions.

Bachelor's degree in Computer Science, Computer Engineering, Electrical Engineering, Mechatronics Engineering, Software Engineering, or a related field.
Practical experience with computer vision and digital image processing.
Experience with Python, especially OpenCV and NumPy.
Experience integrating and executing machine learning and deep learning models.
Experience with PyTorch or equivalent frameworks.
Knowledge of computer vision models for classification, detection, and segmentation.
Experience with ONNX and/or ONNX Runtime.
Knowledge of C/C++.
Experience developing APIs, preferably with FastAPI or equivalent technologies.
Knowledge of Git, automated testing, and collaborative development practices.
Knowledge of software architecture, distributed applications, and Linux.
Experience with Docker and containerized application development.
Experience with cloud environments, preferably AWS.
Preferred experience in industrial computer vision or automated visual inspection systems.
Preferred knowledge of Automated Optical Inspection, real-time video processing, GStreamer, FFmpeg, NVIDIA Jetson, TensorRT, model quantization, pruning, integrated cloud and edge systems, MLOps tools such as MLflow or DVC, CI/CD automation, Kubernetes, C++ computer vision, image-based metrology, and ISO 68-1 and ISO 965 metric thread standards.

1,001-5,000 employees