Coinbase is seeking a Senior Machine Learning Platform Engineer to build the foundational components for feature engineering and training/serving ML models at Coinbase, which are used to combat fraud, personalize user experiences, and analyze blockchains.
Requirements
- You have a strong understanding of distributed systems.
- Experience with the technologies we use (Python, Golang, Ray, Tecton, Spark, Airflow, Databricks, Snowflake, and DynamoDB).
- Experience building ML models and working with ML systems.
- Experience working on a platform team, and building developer tooling.
- You have a mastery of the fundamentals, such that you can quickly jump between many varied technologies and still operate at a high level.
Responsibilities
- Form a deep understanding of our Machine Learning Engineers’ needs and our current capabilities and gaps.
- Mentor our talented junior engineers on how to build high quality software, and take their skills to the next level.
- Continually raise our engineering standards to maintain high-availability and low-latency for our ML inference infrastructure that runs both predictive ML models and LLMs.
- Optimize low latency streaming pipelines to give our ML models the freshest and highest quality data.
- Evangelize state-of-the-art practices on building high-performance distributed training jobs that process large volumes of data.
- Build tooling to observe the quality of data going into our models and to detect degradations impacting model performance.
Other
- 5+ yrs of industry experience as a Software Engineer.
- You lead by example through high quality code and excellent communication skills.
- You have a great sense of design, and can bring clarity to complex technical requirements.
- You treat other engineers as a customer, and have an obsessive focus on delivering them a seamless experience.
- In-person participation is required throughout the year. Team and company-wide offsites are held multiple times annually to foster collaboration, connection, and alignment. Attendance is expected and fully supported.
