Reading the job description

Core details

A JD is a weighted rubric disguised as prose. Staff candidates extract: scope (product surface), scale hints, stack (your fit), culture signals (ownership, 0→1 vs maintenance), and title inflation.

Parse checklist

JD signal	Turn into prep
“Distributed systems”	CAP/PACELC, idempotency, storage engines, outages
“React / Next.js”	RSC, hydration, HTTP cache
“Node.js / TypeScript”	event loop, async, pools, backpressure
“AWS”	RDS, DynamoDB, ElastiCache, IAM mental model
“Kubernetes / Docker”	health probes, rollout, connection fan-out to data tier
“Mentor / lead”	2 behavioral stories with metrics and conflict

Understanding

Interviewers score consistency: if the JD stresses reliability, weak incident + SLO talk hurts even if you ace LeetCode. If it stresses product velocity, show trade-off stories (scope vs quality).

Senior understanding

Trap	Recovery
Stack mismatch	Name transferable depth (“I haven’t shipped X in prod, here’s how I’d ramp + what I’ve done that’s adjacent”)
Title inflation	Ask scope: team size, on-call, blast radius

Diagram

JD text
 ├── Stack keywords → deep pages (/frontend, /backend, /databases)
 ├── Scale keywords → HLD practice list (/dsa/system-design/hld)
 ├── Leadership verbs → STAR stories (/interview/topics/behavioral-stories-structure)
 └── Gaps → honest plan + one proof project or mock