Machine Learning Engineer

Date Opened

09/25/2025

Job Type

Full time

Work Experience

5+ years

Industry

IT Services

City

Львів

State/Province

Lvivska

Country

Ukraine

Zip/Postal Code

79000

We are building an object detection and monitoring solution leveraging SAR data from different providers. Our analytical solution ingests high-resolution SAR imagery, applies advanced classification and sensor-fusion algorithms, and delivers actionable insights to operators. The platform is implemented as a suite of containerized microservices — communicating via Kafka and REST APIs — and orchestrated on Kubernetes for seamless scalability and resilience.

Technology stack & Development tools: Programming & ML Frameworks: Python, PyTorch (Deep Learning); Data & Image Processing: GDAL (geospatial data library), ESA SNAP toolbox (SAR image processing).

• Model Development: Design, implement, and refine deep learning models for object detection.
• Data Pipeline & Preprocessing: Build and maintain data pipelines to ingest and preprocess large images, including steps like calibration, noise reduction, image tiling/patching, and normalization, to prepare data for training and inference.
• Dataset Creation & Annotation: Oversee the creation of annotated datasets for training — work with the team to define annotation guidelines and use tools like CVAT for labeling. Ensure data quality by reviewing annotations and making iterative improvements to the dataset.
• Model Training & Tuning: Train deep learning models on large-scale image datasets using PyTorch. Perform hyperparameter tuning, experiment tracking (with MLFlow or similar), and rigorous evaluation on validation/test sets. Analyze model performance (precision, recall, false alarms) and conduct error analysis to guide improvements.
• Integration & Deployment: Collaborate with software engineers to integrate the trained models into a production inference pipeline. Containerize models (Docker) and optimize runtime (leveraging GPUs) for deployment in the Kubernetes-based microservice architecture. Ensure the inference service communicates correctly via REST APIs or message brokers (Kafka) with the rest of the system.
• Performance Monitoring: Implement monitoring for the ML components in production — track inference speed, resource usage, and detection accuracy drift. Develop strategies for model update/re-training as new data arrives, and troubleshoot issues with the pipeline by refining the model or adding domain-specific post-processing.
• Technical Leadership: Act as a subject matter expert in machine learning within the team — mentor juniors, review code and model designs, and ensure best practices in experiment reproducibility, code quality, and documentation are followed. Work closely with the project's Tech Lead and other stakeholders to align ML development with overall product goals.

• Experience: 5+ years of hands-on experience in machine learning engineering or applied computer vision, with a track record of building and deploying models for object detection or segmentation tasks in production.
• Deep Learning Expertise: Strong proficiency in deep learning frameworks (especially PyTorch) and related libraries (numpy, pandas, OpenCV). Solid understanding of convolutional neural networks and modern CV architectures for object detection (e.g., CNN-based detectors, FPN, Transformer-based models).
• Programming Skills: Excellent Python programming skills and experience with writing clean, efficient code for data processing and model implementation. Familiarity with software engineering practices (version control, code reviews, testing).
  
  Computer Vision & Image Processing: Experience in processing large-scale imagery or remote sensing data. Understanding of image preprocessing techniques (filtering, normalization, augmentation) and ideally some exposure to Synthetic Aperture Radar (SAR) data or other geospatial imagery.
• Data Pipeline & Tools: Experience building end-to-end ML pipelines. Comfortable with tools for managing data and experiments — e.g., using MLFlow for experiment tracking and DVC or similar for dataset versioning. Ability to work with geospatial data libraries (GDAL, Rasterio) and formats.
• Deployment & MLOps: Proven ability to deploy ML models as services. Experience containerizing applications with Docker and knowledge of orchestrating services (Kubernetes) in a production environment. Understanding of how to optimize inference on GPU hardware and monitor model performance over time.
• Problem-Solving: Strong analytical and problem-solving skills, especially in troubleshooting model issues and improving algorithm robustness (for example, reducing false positives in challenging scenarios like coastline or harbor areas).
• Education: Bachelor's or Master's degree in Computer Science, Data Science, Engineering, or a related field. (Advanced experience can substitute for formal degrees.)
• Communication: Good communication skills to convey complex ML concepts to team members and stakeholders. Ability to document workflows and present results clearly

Would be a plus:

• Maritime Domain Knowledge: Understanding of maritime object characteristics or experience in maritime surveillance projects.
  
  Experience with object detection, maritime AIS data, or satellite imagery analytics in a defense or oceanography context.
• Additional Tools: Experience with GIS tools (e.g., PostGIS, QGIS) or other relevant frameworks for geospatial analysis. Familiarity with other deep learning frameworks (TensorFlow) or distributed training frameworks.
• Domain Algorithms: Knowledge of techniques to reduce false detections in marine environments (e.g., land masking, integrating context from other sensors) or previous work on object detection in low-quality/noisy images.
• Leadership: Experience in a senior or lead role on an AI/ML team, driving projects from prototyping to production. This includes experience in mentoring team members and collaborating in an Agile development environment.

Professional development

• Highly experienced professional community: more than 50% employees have a senior level of competence
• Personal development plan and regular Performance Appraisal
• Clear rotation process with opportunity to switch between different roles, projects or technology stack
• Attendance of professional conferences, meetups and certifications (coverage upon on business needs)
• Internal training programs, free SmartTalks and TechTalks inside the company
• Access to Pluralsight
• Corporate English classes

Well-being

• Flexible schedule and possibility to work remotely: feel free to choose an option upon your preferences
• Paid vacation annually