Staff SW Engineer, Machine Learning

BlackSky is a real-time intelligence company that operates a sophisticated space-based intelligence platform, offering satellite imagery and analytics to military and commercial clients. The Staff Software Engineer in Machine Learning will focus on integrating machine learning into automated systems, particularly in the areas of computer vision and natural language processing, while collaborating with the analytics team to enhance existing solutions.

Responsibilities:

• Design and implement solutions for internal and external customers that exploit traditional machine learning and novel deep learning for next-generation satellite imagery analytics
• Plan and conduct research projects related to computer vision, time series analysis, content curation, probabilistic modeling, machine learning, predictive analytics, and geometric modeling
• Develop algorithms, models, and analytical tools for solving domain specific business problems
• Implement production quality analytics and models into the SpectraAI codebase (Python)
• Collaborate with management and technical team on product strategy
• Collaborate with infrastructure developers and machine learning quality engineers to build robust analytics for production use cases
• Independently design and conduct experiments, tests hypothesis, implement model and loss function code, train models, and interpret experiment results following a machine learning process based on high level project objectives
• Other job-related duties as assigned

Requirements:

• At least eight years of hands-on experience as a machine learning engineer or data scientist
• Bachelor's Degree or higher in one of the following fields: computer science, mathematics, physics, statistics, or another computational field with a strong background of using machine learning/data mining for predictive modeling or time series analysis
• Extensive experience developing machine learning based software solutions. In particular, developing models in Python 3, PyTorch, Tensorflow, Keras, or scikit-learn
• Working knowledge of a wide range of machine learning concepts including supervised and unsupervised deep learning methods for both classification and regression
• Experience performing research in both groups and as a solo effort with a history of implementing algorithms directly from research papers
• Experience conducting literature review and applying concepts to programs or products
• Strong ability to communicate concepts and analytical results with customers, management, and the technical team, highlighting actionable insights
• Hands-on experience working with large data sets including data cleansing/transformation, statistical analyses, and visualization (using Python libraries such as Pandas, NumPy, etc.)
• PhD./Master's degree in the previously mentioned fields
• Experience working with remote sensing data, ideally satellite imagery
• Experience with cloud-based MLOps tools such as ClearML, Weights & Biases, Kubeflow, or MLFlow
• Experience working with Kubernetes-based infrastructure
• Experience with tracking and motion detection algorithms
• Experience with maritime data for analysis and modeling
• Experience working with geospatial data and geospatial Python libraries (GDAL, shapely, rasterio, etc.)
• Experience developing asynchronous processing algorithms and Cloud-based solutions (especially AWS services like EC2 & S3)