ML Engineer: Real-Time ETA Systems for Rides

Lyft Inc; San Francisco (CA); Hybrid; USD 141,000 - 176,000; Full time

Benefits offered by this job

Great health insurance and vision

Mental health benefits

Family building benefits

401(k) with company match

Paid time off (holidays, PTO)

Parental leave (18 weeks)

Job summary

Lyft Inc. is hiring a Machine Learning Engineer for the ETA team responsible for estimating/predicting ETAs used in ride matching, pricing estimates, and user experience. The role uses AWS, Kubernetes, Go, Spark, Python, and Apache Airflow across distributed teams.

You will develop ML models, analyze data, and write production-grade code that scales to millions of requests per day, in a hybrid in-office setting in San Francisco.

Qualifications

• 3+ years of machine learning experience.
• BS/MS/PhD in CS or related quantitative field.
• Nice-to-have: Airflow and Spark experience in data pipelines.

Responsibilities

• Perform data analysis and build proof-of-concept to explore and compare ML and non-ML solutions
• Be able to make effective tradeoffs between model accuracy, its productization complexity and runtime performance
• Develop statistical, machine learning, or optimization models
• Write production quality code that can scale well to serve millions of requests per day
• Participate in code reviews, design reviews, production on-call support and incident triaging process.
• Write well-crafted, well-tested, readable, maintainable code

Skills

ML workflow

Data analysis

Production-grade coding

Cross-functional collaboration

Critical thinking

Education

BS/MS/PhD in CS or quantitative field

Tools

AWS

Kubernetes

Go

Spark

Python

Apache Airflow

Job description

Lyft Inc. is hiring a Machine Learning Engineer for the ETA team responsible for estimating/predicting ETAs used in ride matching, pricing estimates, and user experience. The role uses AWS, Kubernetes, Go, Spark, Python, and Apache Airflow across distributed teams.

You will develop ML models, analyze data, and write production-grade code that scales to millions of requests per day, in a hybrid in-office setting in San Francisco.