Machine Learning Engineer

About the Role

Lead the development and deployment of scalable AI solutions for aerial imagery analysis and design of deep learning models capable of analyzing imagery to distinguish between various environmental classes.

Responsibilities

AI Model Development & Optimization

• Design, train, and fine‑tune Convolutional Neural Networks and Vision Transformers for semantic segmentation, object classification, and species identification.
• Build models for identifying vegetation encroachment and develop AI‑driven risk classification and alert systems for vegetation hazards.

Feature Extraction

• Create models tuned to identify lawns, tree canopies, shrubs, hardscapes (driveways, roofs), roadways (dirt, concrete, paved), and powerlines.

Data Pipeline

• Build pipelines to preprocess large datasets of imagery (normalization, tiling, augmentation, masking).
• Establish rules and logic for time‑based change detection and monitoring.
• Implement predictive analysis tools for vegetation growth patterns and maintenance planning.

Optimization

• Balance model accuracy with inference speed to ensure efficient processing of large geographic areas.
• Implement continuous training and model lifecycle management systems, monitor model performance and automate retraining using new datasets.
• Establish version control, evaluation, and deployment procedures for models.

Collaboration & Documentation

• Work closely with GIS analysts, software engineers, product managers, and domain experts.
• Maintain proper documentation of methodologies, workflows, and technical solutions.
• Participate in research, testing, and innovation initiatives.

Qualifications

Experience

• 3+ years in Data Science or Machine Learning with a dedicated focus on Computer Vision.
• Strong understanding and demonstrable experience with various forms of AI.

Proficiency in Python and C++

• Strong coding skills with an emphasis on clean, well‑documented, reusable code.

Deep Learning Frameworks

• Production-level experience with PyTorch (preferred) or Tensor Flow.

Image Analysis

• Deep understanding of CNN architectures such as Deep Lab V3 + and U‑Net.

Preferred Skills

Geospatial Tools

• Experience with satellite or aerial imagery tools (GDAL, ArcGIS, QGIS).

Remote Sensing

• Knowledge of vegetation indices (NDVI) and multispectral imaging.

MLOps

• Experience with Docker, Kubernetes, or AWS Sage Maker.

Cloud Computing

• Familiarity with AWS.