Amazon SDE I / AUTA APJ — Interview study book

Notes for Adarsh Vishwakarma. Java Live Code. Job 10454435 still has no public IE that names a live-round question — this book is not a prediction. Unnamed stays unnamed. R2 lock: 18 Aug 2026. Ignore Round 2 = 14 Aug in older files.

Each chapter keeps a short scan-sheet (bullets) and then a spoken answer in complete sentences with an example. Say the paragraphs, not the telegram list. Full DSA problem cards stay in Answer-BIBLE. Questions are labeled IE-asked Resume-derived Standard CS.

How to use in a 60-minute Live Code

1. LP 10–15 min → chapter 11 (spoken STAR) + chapter 07 project pitches. Companion: Answer-BIBLE §4 / §11.
2. DSA 40–45 min → Cheat-Sheet then Answer-BIBLE §1. This book chapter 01 is DS internals (why a heap), not every LeetCode card.
3. If the slot is LLD → chapters 02 and 06. Rate Limiter is one SDE I Second Technical (OA+3), not UTA two-DSA.
4. If they open GitHub on InstaRecon: security-awareness demo, consent, no production attacks. Then Ylogx / IQVIA / Argus / StratifyLabs. If they open valAgent / KARYA Node / Warpflow: label GitHub in the first sentence — those are not Aug 2026 resume bullets.

Chapters

1. Java data structures — heaps, HashMap, deque, trees, graphs, DSU
2. OOP and OOP LLD — SOLID, locker, bookstore, logger, parking, find
3. Networking and backend — TCP/HTTP/DNS/TLS, REST/GraphQL/ProtoBuf, Nest/FastAPI
4. Auth and OAuth — JWT, OAuth2, RBAC, Ylogx RLS 3 tiers
5. Databases and ER — schema, indexes, BI / RAG / PPE sketches
6. System design — logics + Amazon-style products
7. Resume projects — IQVIA, Ylogx, Horizon, StratifyLabs, GiftedBooks, Argus; GitHub valAgent / KARYA Node / Warpflow
8. CNN and computer vision — YOLO vs OpenCV, Argus metrics
9. LangGraph, LangChain, LLMs — Hybrid RAG, evals, where not to use
10. OS, cloud, question index — threads, ECS, CI/CD
11. Leadership Principles — all 16 official LPs (July 2021 PDF), spoken STAR, IE-asked prompts, proud-of + bugs

Resume metrics (do not invent extras)

• Ylogx: 40% faster reports, 99.9% uptime, SQL RAG +65% productivity, RLS+RBAC 3 tiers, Redis −35% bot DB latency, 30 KPI dashboards +60% ops, sub-210 ms, CloudFront / ECS / Docker / ALB / Route 53.
• IQVIA: LangGraph Deep Research 200+ websites; Hybrid RAG Azure AI Search + GraphDB; 200+ page BRDs; LangSmith.
• Horizon: ERC 2024 17th / 80+, GStreamer 60 FPS, +40% obstacle detection, ZED 2 2M+ pts/s, costmap −55% collision.
• Argus: YOLOv9 73%→89% mAP, 15k images, 24 FPS, −50% violations, 20+ cameras, Postgres logs, 2× compliance.
• GiftedBooks: RAG PDF sub-300 ms, 99.5% uptime, +35% reading, +50% engagement, 2.5× comprehension, doubts hours → 3–10 min.
• StratifyLabs: −30% ML iteration, marketplace 50+ models, WebGL/URDF, Gemini RAG bots.
• Live Code is Java. Internships shipped Python / TypeScript (plus ROS2 on Horizon).