ML Engineer (AI Developer)

Job Description

• Core AI Frameworks: Strong proficiency in Python, PyTorch, and TensorFlow and other Deep Learning libriries for developing denoising and enhancement models.
• Computer Vision & Enhancement: Deep understanding of image transformation matrices and granular control over image enhancement parameters.
• Advanced Image Metrics: Experience in calculating and thresholding quality metrics, including brightness, noise, and shape indices.
• AI-Assisted Labeling: Hands-on experience with Label Studio for real-time interactive AI-assisted segmentation and data annotation.
• MLOps for Vision: Proficiency in MLflow for artifact tracking and DVC for managing large image dataset versioning.
• Orchestration & Tools: Experience using Apache Airflow to automate complex enhancement workflows and OpenVINO for model optimization.
• Statistical Auditing: Ability to perform statistical reviews of evaluation results to ensure production safety and model accuracy.

Job Responsibilities

As an ML Engineer (AI Developer), you will bridge the gap between process engineering and scalable AI infrastructure.

• Modular Image Processing: Design and develop custom image pre-processing flows and device-specific adapters to ensure customer needs are met across different device types.
• Pipeline Engineering: Construct various types of data processing pipelines specific to use cases on CD-SEM devices.
• Deep Learning Implementation: Utilize modern Deep Learning libraries to perform ML model training and evaluation specifically for image enhancement.
• Quality Control & Validation: Implement automated model validation and image enhancement quality control to compare processed vs. original images.
• Algorithm Integration: Add new algorithms and image processing tools via a plug-in approach to the unified platform.

Department/Project Description

Our client is evolving its existing application from a monolithic, device-specific tool into a scalable, unified MLOps platform.

The project focuses on supporting multiple Critical Dimension Scanning Electron Microscope (CD-SEM) devices and expanding AI denoising model usage across diverse environments. You will join a team moving away from hard-coded solutions toward a modular "Plugin and Adapter" architecture that automates the machine learning lifecycle—from data ingestion to edge deployment.