Interview Prep DevNinja
Home

Company prep

Company-specific interview guides

Google, Amazon, Meta, Microsoft, Netflix, Uber, AI labs, and enterprise patterns.

Difficulty Guide Read ~20 min Master per-company 1–2 weeks Levels All

Company guides below are distilled from publicly discussed interview patterns (candidate reports, engineering blogs). They are preparation heuristics, not leaked questions.

Google

Difficulty Hard Read ~5 min Master 2–4 weeks Levels L3–L6
  • Coding: Prefer patterns over memorization — graphs, BS on answer, hard DP. Labs: Islands, Word Ladder, Alien Dictionary, Median/rotated search family.
  • Design: Crawler, Maps, Drive, YouTube, rate limiter, KV. Emphasize scale + freshness + failure.
  • AI roles: RAG eval, serving, retrieval quality.
  • Evaluating: Clarity, generalization, testing mindset.

Amazon

Difficulty Medium→Hard Read ~5 min Master 2–4 weeks Levels SDE I–II / L5–L6
  • Behavioral first: Leadership Principles with STAR — Ownership, Customer Obsession, Dive Deep, Bias for Action, Disagree and Commit. See Behavioral.
  • Coding: Speed + working code. OA favorites: Two Sum family, LRU, islands, rotting oranges, stock, top-k.
  • Design: Shortener, rate limiter, Dynamo-style KV, payments, notifications.

Meta

Difficulty Hard Read ~5 min Master 2–4 weeks Levels E3–E5+
  • Coding: Often two mediums in one round — windows, trees, intervals, heaps. Min window, meeting rooms, serialize tree.
  • Design: News feed, chat, Instagram stories adjacency, counters.
  • Speed + product sense matter.

Microsoft

Difficulty Medium→Hard Read ~4 min Master 2–3 weeks Levels L59–L63+
  • Coding similar to FAANG mediums; design often Teams/Azure flavored: chat, storage, identity.
  • Expect collaboration and debugging narratives.

Netflix

Difficulty Hard Read ~4 min Master 1–3 weeks Levels Senior+
  • Video edge + recommendations + resilience culture (circuit breakers, chaos).
  • Labs: YouTube/video Q6, recsys AI Q6, caching Q9, comparisons CDN/cache.

Uber / Lyft

Difficulty Hard Read ~4 min Master 1–3 weeks Levels Senior+
  • Geo, marketplace dispatch, ETA, city sharding — SD Q5. Kafka-heavy data paths Q16.

OpenAI-adjacent / AI platforms

Difficulty Hard Read ~4 min Master 2–4 weeks Levels Mid–Staff
  • Full AI Lab: RAG, agents, eval, serving, tenancy, security.

Red Hat / enterprise open source

  • Expect Linux/Kubernetes, operators, observability, and design for on-prem constraints.
  • Map to scaling, queues, and reliable upgrades — emphasize operable systems.