Machine Learning Engineer

Machine Learning Engineer II at Scribd, Inc. – design, build, and optimize high-impact ML systems that serve millions of users in near real time. You will work on projects spanning core ML platform improvements and integrating models into the product experience.

At Scribd (pronounced “scribbed”), our mission is to spark human curiosity. Join our team as we create a world of stories and knowledge, democratize the exchange of ideas and information, and empower collective expertise through our three products:
Everand, Scribd, and Slideshare.

We support a culture where employees can be real and bold; where we debate and commit as we embrace plot twists; and where every employee is empowered to take action as we prioritize the customer. Scribd Flex allows employees to choose the daily work style in partnership with their manager; occasional in-person attendance is required for all Scribd employees, regardless of location.

We hire for “GRIT” (Goals, Results, Innovation, Team) and expect that mindset in how we work.

Our Machine Learning team builds platform and product applications powering personalized discovery, recommendations, and generative AI features across Scribd, Slideshare, and Everand. The team maintains the Orion ML Platform—providing core ML infrastructure, including a feature store, model registry, model inference systems, and embedding-based retrieval (E ). The MLE team collaborates with Product to deliver zero-to-one integrations of ML into user-facing features like recommendations, near-real-time personalization, and AskAI LLM-powered experiences.

We are seeking a Machine Learning Engineer II to design, build, and optimize high-impact ML systems that serve millions of users in near real time. You will work on projects spanning improving our core ML platform to integrating models into the product experience.

Our Machine Learning team uses a range of technologies to build and operate large-scale ML systems. Our regular toolkit includes:

• Languages:
  
  Python, Golang, Scala, Ruby on Rails
• Orchestration & Pipelines:
  Airflow, Databricks, Spark
• ML & AI: AWS Sagemaker, embedding-based retrieval (Weaviate), feature store, model registry, model serving platforms, LLM providers like OpenAI, Anthropic, Gemini
• APIs & Integration: HTTP APIs, gRPC
• Infrastructure & Cloud: AWS (Lambda, ECS, EKS, SQS, Elasti Cache, Cloud Watch), Datadog, Terraform

• Design, build, and optimize ML pipelines, including data ingestion, feature engineering, training, and deployment for large-scale, real-time systems.
• Improve and extend core ML Platform capabilities such as the feature store, model registry, and embedding-based retrieval services.
• Collaborate with product software engineers to integrate ML models into user-facing features like recommendations, personalization, and AskAI.
• Conduct model experimentation, A/B testing, and performance analysis to guide production deployment.
• Optimize and refactor existing systems for performance, scalability, and reliability.
• Ensure data accuracy, integrity, and quality through automated validation and monitoring.
• Participate in code reviews and uphold engineering best practices.
• Manage and maintain ML infrastructure in cloud environments, including deployment pipelines, security, and monitoring.

Must Have

• 3+ years of experience as a professional software or machine learning engineer.
• Proficiency in at least one key programming language (preferably Python or Golang; Scala or Ruby also considered).
• Hands-on experience building ML pipelines and working with distributed data processing frameworks like Apache Spark, Databricks, or similar.
• Experience working with systems at scale and deploying to production environments.
• Cloud experience (AWS, Azure, or GCP), including building, deploying, and optimizing solutions with ECS, EKS, or AWS Lambda.
• Strong understanding of ML model trade-offs, scaling considerations, and performance optimization.
• Bachelor’s in Computer Science or equivalent professional experience.

Nice to Have

• Experience with embedding-based retrieval, recommendation systems, ranking models, or large…