Netflix

Software Engineer L5/L6 — Model Evaluations & Data Curation (MEDC)

Netflix
Apply
2 days ago
Remote, United StatesSenior / Staff+
H1B sponsor

Base Salary

$600k - $1066k/yr

Responsibilities

  • Design and build reusable data-curation infrastructure, libraries, and workflows that replace ad-hoc notebooks and accelerate productionization.
  • Build scalable LLM-driven data-transformation pipelines for creating training and evaluation data from sources such as the Netflix catalog and metadata.
  • Develop sampling strategies based on coverage, diversity, difficulty, and content or member-segment balance.
  • Develop filtering and quality-control methods including LLM-as-judge scoring, evaluation-model scoring, deduplication, and validation.
  • Partner with researchers to measure how curation choices affect model performance and connect data-quality signals to model outcomes.
  • Create discoverable, versioned, lineage-tracked, and reproducible dataset artifacts.
  • Drive adoption of shared data-curation practices across MEDC and partner modeling teams.

Requirements

  • Strong Python software engineering experience building reusable infrastructure, libraries, or frameworks for engineers and researchers.
  • Experience building LLM-driven data-generation or transformation pipelines, including synthetic data, structured outputs, or large-scale batch inference.
  • Hands-on experience with sampling, filtering, deduplication, and model-based data-quality scoring such as LLM-as-judge.
  • Understanding of how data choices affect model behavior and ability to design experiments measuring those effects.
  • Experience with distributed data processing such as Spark, Ray, or similar systems.
  • Strong collaboration skills with researchers, data scientists, and platform teams.
  • For L6, experience with LLM evaluation systems and technical leadership setting direction for multi-engineer data and evaluation infrastructure.
  • Preferred experience with dataset versioning, lineage, artifact management, large-scale LLM inference optimization, human annotation workflows, LLM-judge calibration, pipeline orchestration, recommendation systems, personalization, search, or content catalogs and metadata.

Benefits

  • Comprehensive health plans and mental-health support.
  • 401(k) retirement plan with employer match, stock option program, disability programs, health savings and flexible spending accounts, family-forming benefits, and life and serious injury benefits.
  • Paid leave programs; full-time salaried employees are immediately entitled to flexible time off.
  • Full-time hourly employees accrue 35 days annually for vacation, holidays, and sick paid time off.

Tech Stack

Apache AirflowApache SparkPython

Categories

Data EngineeringML Engineering
Netflix

About Netflix

10,000+ employees

Netflix builds and operates a global streaming service for TV series, films, games, and live programming, and produces original content through its in-house studio. It serves consumers in 190+ countries via subscription and ad-supported plans on internet-connected devices. Founded in 1997 and headquartered in Los Gatos, California, Netflix is a public company traded on NASDAQ (NFLX).

Contact me