About the teamPremium is the Scribd Inc. pillar behind two reading products: Everand, the audiobook and ebook subscription app, and Fable, the social app where readers track, discuss and discover books. Helping readers find their next book is central to both, so search and recommendations sit at the heart of the products.
You'll join Premium's new Discovery squad. It's a small, dedicated group of recommendations, search, applied research and ML engineers working full time on discovery for Everand and Fable. The squad works closely with Orion, Scribd's ML platform team. Orion runs the feature store, model registry, embedding-based retrieval and model serving behind discovery across Scribd, Slideshare and Everand.
Role overviewWe're looking for a Senior Machine Learning Engineer to own recommendations for Fable and grow into the technical lead for Premium recommendations.
Fable's recommender today is a collaborative-filtering model built in house. It has served readers well, but it has no real-time signals and no learned ranking, and it can't recommend to new or lapsed readers. Your first big project is to move Fable onto Orion, so Fable and Everand share a modern recommendation stack.
Key responsibilities- Lead the migration of Fable's recommendations onto Orion, covering data, features, training, serving and experiments, while keeping today's recommendations running.
- Extend Orion's rankers, feature store and embedding-based retrieval to support Fable, in partnership with the Orion ML platform team.
- Own the health of Fable's recommendation systems in production: quality, latency, cost and on-call.
- Bring Fable's social signals (shelves, ratings, reviews, book clubs, reader similarity) into recommendations for both Fable and Everand.
- Design and analyze A/B tests with product managers and analysts to decide what ships.
- Act as technical lead for Premium recommendations. Shape the roadmap with product managers, and see work that depends on other teams' systems through to delivery.
- Raise the bar on engineering practice in the squad through design reviews, code quality and operational readiness.
Tech you'll work withYou'll work across two stacks: Fable's on Google Cloud and Orion's on AWS.
- Fable: Python (FastAPI, Django), PostgreSQL, BigQuery, Kubernetes, Google Cloud including Vertex AI
- Orion and Everand: Python, Go and gRPC, Ruby on Rails, Databricks and Spark, Airflow, AWS SageMaker, Weaviate for embedding-based retrieval, feature store and model registry
- Infrastructure: AWS (EKS, ECS, Lambda, SQS, ElastiCache), Datadog, Terraform
RequirementsMust have- 5+ years as a professional ML or software engineer, with a track record of shipping production ML systems.
- Hands-on experience building and running recommender systems in production, such as collaborative filtering or matrix factorization, two-tower or embedding retrieval, and learned ranking.
- Experience with large-scale data processing (Spark, Databricks, BigQuery or similar).
- Experience on at least one major cloud (AWS or GCP), including deploying services on containers (Kubernetes, ECS or similar).
- Experience designing and analyzing online experiments.
- A record of leading projects that depend on other teams, and of working directly with product managers.
- Bachelor's or Master's degree in Computer Science, or equivalent professional experience.
Nice to have- Google Cloud experience, especially Vertex AI and BigQuery.
- Experience migrating or replacing a production ML system while keeping it running.
- Social or graph-based recommendations, such as user similarity, follows or groups.
- Search and semantic retrieval (Elasticsearch, Vertex AI Search, vector databases).
- Experience on consumer apps for reading, media or social.
At Scribd, Inc. your base pay is one part of your total compensation package and is determined within a range. Our pay ranges are based on the local cost of labor benchmarks for each specific role, level, and geographic location. San Francisco is our highest geographic market in the United States. In the state of California, the reasonably expected salary range is between $170,000 [minimum salary in our lowest geographic market within California] to $265,000 [maximum salary in our highest geographic market within California].
In the United States, outside of California, the reasonably expected salary range is between $140,000 [minimum salary in our lowest US geographic market outside of California] to $252,000 [maximum salary in our highest US geographic market outside of California].
In Canada, the reasonably expected salary range is between 179,000 CAD[minimum salary in our lowest geographic market] to 237,500 CAD[maximum salary in our highest geographic market].
We carefully consider a wide range of factors when determining compensation, including but not limited to experience; job-related skill sets; relevant education or training; and other business and organizational needs. The salary range listed is for the level at which this job has been scoped. In the event that you are considered for a different level, a higher or lower pay range would apply. This position is also eligible for a competitive equity ownership, and a comprehensive and generous benefits package.
Working at Scribd, Inc.Are you currently based in a location where Scribd, Inc. can employ you? Employees must have their
primary residence in or near one of the following cities. This includes surrounding metro areas or locations within a typical commuting distance:
United States: Atlanta | Austin | Boston | Dallas | Denver | Chicago | Houston | Jacksonville | Los Angeles | Miami | New York City | Phoenix | Portland | Sacramento | Salt Lake City | San Diego | San Francisco | Seattle | Washington D.C.
Canada: Ottawa | Toronto | Vancouver
Mexico: Mexico City
Benefits at Scribd, Inc.- Scribd Flex (flexible work model)
- Comprehensive health, dental, and vision coverage
- Mental health support and disability coverage
- Generous paid time off, including vacation, sick time, holidays, winter break, volunteer time, and sabbaticals
- Paid parental leave and family support benefits
- Retirement matching and employee equity
- Learning and development programs and professional growth opportunities
- Wellness and home office stipends
- Complimentary access to the Scribd, Inc. suite of products
- Enterprise access to leading AI tools
We want our interview process to be accessible to everyone. You can inform us of any reasonable adjustments we can make to better accommodate your needs by emailing [redacted] about the need for adjustments at any point in the interview process.
If you apply for a job with Scribd, Inc. or otherwise engage with us in connection with employment (including as an employee, contractor, or other personnel), the personal information we process in that context is subject to our Employee and Applicant Privacy Policy, which is available here.
