Machine Learning Scientist

San Francisco, open to remote /
Analytics /
Full Time
Space is in our DNA. Orbital Sidekick (OSK) has a mission to leverage the untapped power of hyperspectral imagery (HSI) to address global environmental, health, and safety needs while helping companies reach their sustainability goals. Orbital Sidekick is establishing a space-based infrastructure of HSI sensors and analytic capabilities to enable real-time monitoring, detection, and risk management.

We’re looking for a Machine Learning Scientist who will support our analytics team. This person will work across operations, engineering, product, and analytics teams to help us solve some of the hardest problems in remote sensing.

What you'll do

    • Visualizing meaningful relationships between hyperspectral imagery (HSI)/other geospatial data and measurable/observable phenomenologies
    • Implementing and developing ML models using hyperspectral data Keeping up with the latest ML literature, designing and tracking experiments, tuning models and increase performance
    • Working with engineering, product, and other analytics teams to establish and maintain data exploitation processes to generate reliable industry end-user products
    • Developing algorithms to address problems of change detection and target detection, deriving insights from high dimensional data  
    • Working with operations to design successful field experiments that yield high quality data

What you'll need to succeed

    • Must be US citizen or permanent resident
    • A Bachelor's, Masters, or PhD in CS, EE, Physics, Mathematics, or related
    • 5+ years experience building machine learning models for computer vision, signal processing, or similar applications
    • Python, Julia, MATLAB, C++ or other programming experience, git, AWS, Sagemaker 
    • Experience building models with major deep learning frameworks, e.g. Pytorch, Keras, Tensorflow
    • Basic understanding of estimation and detection theory
    • Knowledgeable about model compression techniques and experience deploying models to both embedded systems and the cloud
    • Understand the challenges of training models on remotely sensed data including noisy data, limited or inaccurate labels, etc.
    • Willingness to wrangle, gather, clean, and build data pipelines
    • Excellent written and oral communication skills
    • Although not a requirement, bonus points if you have:
    • Experience in different aspects of imaging system design and simulation: Optical, Electrical, and Algorithms for Image Processing.
    • Experience with image/hyperspectral image denoising strategies, K-SVD, Minimum Noise Fraction, Wavelet denoising, PCA

What we offer

    • Competitive compensation and equity packages
    • Comprehensive health plan including health/dental/vision/wellness
    • Disability and Life insurance
    • 401K program
    • Flexible time off
    • Commuter benefits
    • Learning and development opportunities
    • Monthly stipend for internet/cell phone/office expenses
    • Flexible in office schedule with lunch Tuesday to Thursday in the San Francisco headquarters
    • Off-sites and team-building activities
    • Remote friendly (we hire in-office and remote employees in the United States)
Orbital Sidekick is committed to building a community where everyone belongs and we invite people from all backgrounds to apply. We are an equal opportunity employer committed to providing employment opportunities regardless of race, religious creed, color, national origin, ancestry, physical disability, mental disability, medical condition, marital status, sex, sexual orientation, gender, gender identity, gender expression, pregnancy, age, military or veteran status, or any other protected classification, in accordance with applicable federal, state, and local laws.