computer vision scientist, generative ai - Hybrid or Remote

about the computer vision scientist, generative ai role:

soona is looking for a computer vision scientist with generative AI expertise to spearhead the machine learning efforts leveraging our extensive library of proprietary images and videos. the potential applications are many, but we are particularly focused on building computer vision and generative models, with an emphasis on diffusion models, to automate content generation and optimize our creative efforts based on our customers' product images. to be successful in this role, you will excel at training computer vision models in areas such as image segmentation, object identification, and adapting diffusion model architectures to generate images with embedded customer products. these models will be deployed to a performant production environment directly accessible by our customers.

about soona:

soona makes it possible for brands to create professional photo and video starting at $39. our studios give customers a playground for creating their content and our online platform makes it possible for any product company in the world to experience a remote shoot. we are creating a fast casual content revolution!

soona is currently supporting a US remote work environment for this role with opportunity for a flex hybrid work environment within our operating cities--Denver and Minneapolis, if that’s your thing.

about tech at soona:

at soona, we’re focused on building a world-class engineering and data organization. we’re developing a highly-scalable platform for real-time customer engagement with our studio creatives and technology that optimizes the content they create. our typical engineering and data projects blend SaaS with e-commerce, providing opportunities to work on everything from app engineering and cloud/server architecture to computer vision and logistics/routing optimization. our tech stack consists primarily of ruby on rails, javascript vue, and python. we pride ourselves on our culture of innovation, community engagement, technical mentorship, and caring for the individual.

our hiring philosophy:

at soona, we look for representation across all intersectionalities of identities, specifically within underrepresented groups. it is these differences that push us towards innovation, curiosity, and success in our business. we believe in providing equal employment opportunities without regard to race, color, religion, age, sex, national origin, disability status, protected veteran status, genetics, sexual orientation, gender identity or expression, or any other characteristic protected by laws or regulations in the locations we operate. this means that timelines of processes may be impacted, depending on our applicant pools.

Requirements

an ideal candidate can:

• build computer vision models to analyze and optimize our vast library of visual assets
  
• adapt generative AI (e.g. diffusion model) architectures to suit specific needs, such as embedding customer products within generated images
• deploy models on distributed GPU instances to serve up near real-time insights in our production image capture environment
• create metadata capture apps to build robust training datasets using our visual assets
• think for themselves and discover new and insightful ways to solve difficult problems
• deliver quality code in an agile framework that ships to a production environment
• communicate with data and engineering teams as well as business stakeholders

has experience in:

• effectively communicating and coding in a remote work environment
  
• python – the core language of the data organization
• training machine learning models, specifically leveraging transformers and neural networks (or similar) for computer vision and generative AI applications
• customizing and deploying diffusion model architectures to meet specific needs
• leveraging GPUs/CUDA vis-à-vis pytorch (or similar)
• aws or equivalent cloud environment
• one or more of: image segmentation and foreground/background isolation, object identification and classification, image quality characterization and enhancement, generative adversarial networks (GANs) and diffusion models for content generation
• kubernetes, docker, terraform, flask/fastAPI (preferred)
• working in a startup environment (preferred)

Benefits

we can offer:

• strong starting salary: $190,000 - $220,000
  
• stock options in a booming startup
• benefits & perks + unlimited pto + intentional culture
• really badass headshots