Staff Applied Scientist

Lyft Lyft · Consumer · San Francisco, CA · Central Market Management & AI

Staff Applied Scientist at Lyft focused on developing and deploying ML and optimization models for pricing and ETA decisions, aiming to scale solutions to millions of calls per day and impact marketplace and rider experience. The role involves production-grade code, custom method development, evaluation against business goals, and establishing monitoring metrics.

What you'd actually do

  1. Write production quality code. Design, build and deploy production-grade ML and Optimization models. Able to build custom methods and tooling beyond off-the-shelf libraries.
  2. Perform data analysis and build proof-of-concepts to explore and propose ML and Optimization solutions to both new and existing problems.
  3. Evaluate machine learning systems against business goals. Collaborate with Engineers to implement algorithms in live systems and ensure the robustness of the systems
  4. Establish metrics and development measurement methodologies to monitor the health of our products, as well as the impacts on user and marketplace outcomes
  5. Drive collaboration and coordination with cross-functional teams

Skills

Required

  • Python
  • production coding environment
  • machine learning methodologies
  • building and evaluating optimization or machine learning models
  • verbal and written communication skills
  • collaborating with others to solve a problem

Nice to have

  • Operations Research
  • Statistics
  • Computer Science
  • custom methods and tooling beyond off-the-shelf libraries

What the JD emphasized

  • production quality code
  • production-grade ML and Optimization models
  • build custom methods and tooling
  • evaluate machine learning systems against business goals
  • implement algorithms in live systems
  • ensure the robustness of the systems
  • establish metrics and development measurement methodologies
  • monitor the health of our products
  • impacts on user and marketplace outcomes

Other signals

  • productionalize pipelines
  • scale to millions of calls per day
  • impact on the marketplace
  • impact on rider experience
  • production quality code
  • deploy production-grade ML and Optimization models
  • build custom methods and tooling
  • evaluate machine learning systems against business goals
  • implement algorithms in live systems
  • ensure the robustness of the systems
  • establish metrics and development measurement methodologies
  • monitor the health of our products
  • impacts on user and marketplace outcomes