Machine Learning Engineer, Relevance and Personalization

Airbnb · Consumer · United States · Software Engineering

Machine Learning Engineer focused on search and recommendation at Airbnb, developing end-to-end ranking algorithms and ecosystems for optimizing business objectives. The role involves building AI technologies across the product stack, from data pipelines to serving, and productionizing ML models at scale.

What you'd actually do

  1. Work with large scale structured and unstructured data, build and continuously improve cutting edge Machine Learning models for Airbnb product, business and operational use cases.
  2. Work collaboratively with cross-functional partners including software engineers, product managers, operations and data scientists, identify opportunities for business impact, understand, refine, and prioritize requirements for machine learning models, drive engineering decisions, and quantify impact.
  3. Hands-on develop, productionize, and operate Machine Learning models and pipelines at scale, including both batch and real-time use cases.
  4. Leverage third-party and in-house Machine Learning tools & infrastructure to develop reusable, highly differentiating and high-performing Machine Learning systems, enable fast model development, low-latency serving and ease of model quality upkeep.

Skills

Required

  • New grad Ph.D in ML/AI or 2+ years of industry experience in applied ML/AI with a M.S. or B.S degree.
  • Strong programming (Scala / Python / Java / C++ or equivalent) and data engineering skills.
  • Deep understanding of Machine Learning best practices (e.g. training/serving skew minimization, A/B test, feature engineering, feature/model selection), algorithms (e.g. neural networks/deep learning, optimization) and domains (eg. natural language processing, computer vision, personalization, search and recommendation, marketplace optimization, anomaly detection).
  • Proven ability to choose the right ML method to solve the problem within current constraints while having a clear vision of the next iterations and a good balance between exploration and exploitation of different techniques.
  • Ability to go deep and build the most impactful solutions while also leading multiple directions across multiple teams and organizations to ensure the success of our mission.

Nice to have

  • Exposure to 3 or more of these technologies: Tensorflow, PyTorch, Kubernetes, Spark, Airflow (or equivalent), Kafka (or equivalent), data warehouse (eg. Hive).

What the JD emphasized

  • end-to-end ranking algorithms
  • end-to-end search ranking product stack
  • low-latency serving

Other signals

  • develop end-to-end ranking algorithms and ecosystems
  • optimize multiple critical business objectives
  • build cutting-edge AI technologies across the end-to-end search ranking product stack
  • develop reusable, highly differentiating and high-performing Machine Learning systems
  • enable fast model development, low-latency serving and ease of model quality upkeep