Senior Applied AI/ML Scientist - Listing Quality
About this role
Resourceful practitioners who enjoy tackling unstructured data challenges and driving end-to-end model development with limited supervision will find a natural fit within this group. The Listing Quality team at Faire focuses on the images, titles, descriptions, and structured attributes that power millions of products across hundreds of thousands of independent brands in a global wholesale marketplace.
Team members work extensively with multi-modal deep learning and large language models to extract product attributes, detect catalog-scale quality issues, and improve imagery through hero selection and cropping. Because listing quality directly impacts search and discovery surfaces, this work serves as one of the highest-leverage components of the platform, requiring close collaboration across product, engineering, design, and analytics functions.
What you will do
- Own applied machine learning projects end-to-end, covering problem framing, model building, shipping, and impact measurement.
- Utilize multi-modal deep learning and large language models to understand catalog content and detect quality issues at scale.
- Enhance product imagery through hero selection, cropping, and ordering so retailers see the best possible representation.
- Construct ranking and exploration methods, including bandit-style selection, to learn which content performs best for specific audiences.
- Refine listing text fields like titles and descriptions while tracking the downstream effect on discovery and conversion rates.
- Establish LLM-as-judge and human-in-the-loop evaluation systems to meet measurable accuracy standards before production deployment.
- Collaborate across product, engineering, design, and analytics teams to turn models into shipped products and business results.
What you need
- Have at least 3 years of industry experience utilizing machine learning to solve real-world problems.
- Bring practical background in relevant business domains like e-commerce, marketplaces, catalog quality, search, or personalization.
- Demonstrate technical proficiency in deep learning, large language models, computer vision, information extraction, or experimentation.
- Possess strong programming skills along with a willingness to learn new tools and techniques.
- Show the capability to drive projects independently and lead model development with limited supervision.
- Exhibit strong communication skills necessary for working within a highly cross-functional team environment.
Nice to have
- Master's or PhD in Computer Science, Statistics, or related STEM fields
- Previous experience with catalog quality, product attribute extraction, or computer vision for e-commerce imagery
- Experience with search and discovery for a two-sided platform
- Experience building and validating LLM evaluation pipelines including prompt iteration and human-in-the-loop workflows
Benefits
- Equity
- Comprehensive benefits designed to support life inside and outside of work
Skills and tech stack
Questions candidates ask
What is the remote work policy?
Employees currently go into the office 3 days per week on Tuesdays, Thursdays, and a third flex day of their choosing. Additionally, hybrid in-office roles have the flexibility to work remotely up to 4 weeks per year.
What is the salary range for this role?
The pay range for this role is $211,000 to $290,500 per year based in San Francisco. Actual base pay is determined by permissible factors such as transferable skills, work experience, market demands, and primary work location.
What kind of data will I work with?
You will work primarily with unstructured data using multi-modal deep learning and LLMs. This involves handling catalog content such as images, titles, descriptions, and structured attributes across millions of products.
Who will I be collaborating with?
You will partner across product, engineering, design, and analytics teams. You will work alongside experienced Applied AI/ML Scientists from companies like Uber, Airbnb, Square, Facebook, and Pinterest.