Machine Learning Engineer

Summary

Join EDO, the TV outcomes company, as a Machine Learning Engineer focused on performance optimization. You will be part of a team improving modeling pipelines for our Ad EnGage product, working with massive datasets using Python, DBT, and Snowflake. Your responsibilities include identifying and resolving performance bottlenecks in queries and code, optimizing data processing, and developing profiling test suites. This remote position requires 5+ years of experience in machine learning engineering or related fields, including proficiency in Python, SQL, and columnar databases. EDO offers a competitive compensation package including equity, flexible time off, comprehensive health benefits, and a 401(k) plan.

Requirements

• A minimum of 5 years of machine learning engineering, software engineering, or data science experience, including at least 3 years of development/maintenance of machine learning tools or systems
• A self-driven individual who can own projects and communicate across technical teams to determine where and how to resolve performance issues
• Strong knowledge of columnar databases (such as Snowflake, Redshift, etc), high performance data processing in python (like Polars or DuckDB), and Python memory and runtime optimization
• Strong proficiency with Python, especially memory and CPU profiling
• Strong proficiency in SQL, including inspection of query plans, query optimization, and strategies for working with slowly changing dimension tables
• A strong understanding of software engineering practices, principles, and fundamentals

Responsibilities

• Identify performance bottlenecks in Snowflake queries, and either tune modeling code to query data more efficiently or work with data engineering teams to change upstream tables for more efficient reads
• Identify memory inefficiencies in Python code (ex. memory leaks, excessive copies) and rewrite more efficiently
• Identify processing inefficiencies (missing parallelization, repeated computation, etc.) in data processing, modeling, and scoring code
• Identify inefficient Pandas operations and optimize or migrate the step to Polars or other alternatives
• Diagnose out-of-memory errors in modeling pipelines
• Develop runtime and memory profiling test suites to detect performance regressions in our modeling codebase

Preferred Qualifications

• SQLalchemy and Snowflake-specific expertise
• Memory and CPU profiling test suites
• Development and test pipelines, continuous integration
• Development of feature engineering pipelines and DBT
• Directly working with vendor engineers to resolve performance regressions

Benefits

• Mid-stage equity and competitive salary
• Flexible Time Off
• Medical, dental and vision coverage, deeply discounted by EDO
• 401(k) plan, FSA, HSA
• Commuter Benefits
• When in an office, employee meals, snacks, and more