← All guides

How I’d prep for Netflix’s Full-stack Engineer, User Data & Consent Experiences role in 30 days

Updated Nov 19, 2026

The job

Full-stack Engineer 4, User Data & Consent Experiences, Netflix

Remote (US), or hybrid in Los Gatos · $250K – $413K (salary; you choose your salary and stock option mix) · Posted Nov 16, 2026

View the posting →

What this team actually does

This team sits in Netflix’s Identity Engineering group, which owns login, profiles, access control and account settings. This team builds the screens where Netflix asks for your consent and collects your data: “1P Data Collection, Privacy Consent Collection, across Netflix products,” which the posting lists as the streaming apps, Games, Live, Ads and Netflix House.

The posting is clear about how they build it: “Using Server-driven Web UI technologies, you’ll create modular experiences that minimize client-side logic,” with “server driven UI, javascript and graphQL technologies.” You’d also “lead technical direction for Server Driven UI components.” And because it’s consent, every screen has to balance “business objectives, regulatory requirements” and a good experience, in every country Netflix runs in.

AreaWhat to know
Front endReact and TypeScript on the web
Back endJava and Spring Boot. Netflix built its own open source GraphQL framework for Spring Boot, called DGS
APIGraphQL (plus REST). Apps fetch server-driven screens over GraphQL
UI architectureServer-driven UI: the server describes the screen, and each app (TV, phone, web) draws it natively
ClientsWeb, TV, iOS (Swift) and Android (Kotlin)
DomainConsent and data collection under GDPR and CCPA, run as A/B experiments
The stack, from the posting, older postings for the same team, and Netflix’s own conference talks.

How I found the stack (do this for any job)

  1. Read the posting closely. It says “TypeScript/React and Java/Springboot” and “GraphQL, server-driven UI techniques, and RESTful APIs.” Server-driven UI is the part most candidates won’t know, so that’s where I dug.
  2. Read the older postings for the same team. An older Full-stack Engineer 5 posting has the same mission text word for word, plus “Lead technical direction for Server Driven UI components, APIs and GraphQL schema design” and “Javascript, React, Java, GraphQL.” A sibling posting adds “client-friendly schemas” and “backend microservices & APIs.”
  3. Find the talks. In May 2026, a Netflix engineer gave a talk at GraphQLConf, “Screens on Shuffle: How Netflix Scales Server-Driven, Ever-Changing Pages”. The slides are public. The short version: “Server describes the page. Client renders it.” Each app tells the server what it can render, and the server returns only that, so pages change without app releases. An older 2024 talk from Netflix’s growth engineering team shows the same idea for the pop-up messages Netflix shows on every device.
  4. Find the framework. Netflix open-sourced DGS, its framework for building GraphQL services with Spring Boot, in 2021 (InfoQ coverage). If you’re building GraphQL in Java at Netflix, this is the tool to learn.
  5. Look up the people. Search LinkedIn (through Google) for Netflix engineers who list Identity Engineering, consent or server-driven UI, and read what they worked on.

So: React and TypeScript, Spring Boot with DGS, GraphQL schema design, and server-driven UI, all applied to consent and privacy. That’s where I’d spend the 30 days.

Understand the product first

  1. Open netflix.com in a private window and look at the cookie and privacy choices it gives you. Note what’s on by default, what you can turn off, and how many clicks it takes.
  2. Sign in and go through your account settings for anything about privacy, data or communications. Write down every choice Netflix lets you make about your data.
  3. Watch for the pop-ups Netflix shows across your TV, phone and browser (new terms, plan changes, payment updates). Notice that the same message looks native on each device. That’s server-driven UI at work.
  4. Read the Screens on Shuffle slides end to end. Sketch the page → section → item structure, and the “capabilities” each app sends with its request.

Then ask yourself the questions this team answers every day: what does a user in Germany see that a user in California doesn’t? What happens when a TV app from three years ago gets a new kind of consent screen? If someone withdraws consent, how fast does every other Netflix system find out? How do you prove what a user agreed to, and when? How do you A/B test wording on a legal screen without breaking the law?

System design questions

Netflix’s system design round is widely described as one of the hardest (see interviewing.io and Exponent). Reported prompts include designing Netflix itself, an API rate limiter and an internal event logging system. These are the ones I’d expect for this team:

  • Design server-driven UI for Netflix’s consent screens across TV, web, iOS and Android. How do you ship a new component without breaking apps that are years old?
  • Design a consent service that every Netflix app and backend checks. Reads happen at every login and must be fast; writes must never be lost; every change must be auditable.
  • A user withdraws consent for personalized ads. Design how that change reaches the ads system, the data pipelines and every cache, and how you prove it did.
  • Design a system that shows in-app messages (new terms, a payment problem) on every device, with targeting, frequency limits and A/B tests.
  • Design the GraphQL API for a server-driven page. How do you model components, avoid N+1 queries, paginate, and deprecate fields safely?
  • Design an experimentation platform: assign users to variants consistently across devices, log exposures, and measure the result.
  • Design an API rate limiter, and an internal event logging system (both reported from Netflix interviews).

Concepts they’ll probe

  • GraphQL schema design: nullability, interfaces and unions, input types, connection-style pagination, and deprecating fields instead of versioning.
  • GraphQL performance: the N+1 problem, DataLoader batching, query cost limits, persisted queries and caching.
  • Federation: splitting one graph across many services, and who owns which type.
  • Server-driven UI: coarse-grained (“server decides what, client decides how”) vs fine-grained vs a thin client; component registries; fallbacks for unknown components; client capabilities.
  • Backward and forward compatibility: old apps reading new responses, new apps reading old ones, and rolling out schema changes safely.
  • Spring Boot: dependency injection, configuration, Spring Security, testing, and talking to Postgres.
  • React and TypeScript: rendering from data, discriminated unions for component types, forms and accessibility.
  • Consent and privacy law basics: GDPR consent must be freely given, specific, informed and as easy to withdraw as to give; CCPA’s right to opt out of the sale or sharing of data; Global Privacy Control.
  • Data modeling for audits: append-only consent records, policy versions, and proving what someone agreed to and when.
  • Experiments: stable assignment, exposure logging, guardrail metrics, and what you can’t test on a legal screen.
  • Product engineering: loading, empty and error states, and measuring business impact (the posting stresses this).

Unlock the rest

  • The week-by-week 30-day plan, with daily practice on Devmaxx
  • Exactly what to read each week, from the DGS docs to Production Ready GraphQL
  • The project: a server-driven consent flow with Spring Boot, DGS, GraphQL, React, TypeScript and Postgres, with GDPR and CCPA versions and an A/B test
  • The level bar: what Netflix expects from a Full-stack Engineer 4, and how to prep for the culture interview
  • The resource list with every link

Get the full guide, free

Enter your email and the guide opens right here. You’ll also get the occasional Devmaxx email. Unsubscribe anytime.

Netflix Full Stack Engineer Interview Prep: A 30-Day Plan · Devmaxx