Stop regretting and start working.

ANZINKKKKKK FUCKKKKKKKKKKKKKKKKKKKK

You are testing a caching layer (Redis). You request a key that doesn't exist. The cache misses. The app queries the database. The database returns 'Not Found'. The app does *not* cache 'Not Found'. An attacker sends 1 million random keys. Every request bypasses the cache

Kenapa gw liatnya muka si dia 😩😩😩

Ah mungkin hanya perasaanku saja

X goes down for 30 minutes and you can feel it. 1. “Infra team, please triage.” 2. “Which infra team?” 3. “The one that built it in 2018.” 4. “Yeah, they left.” I feel that X lost the people who knew where those weird cron job and even more complex playbooks live!

You're in a FAANG Backend engineer interview. They ask: "Design Twitter's (X) trending algorithm for 500 million users." Here's the breakdown:👇👇

CQRS quick notes (system design): 1. Split models: Commands write. Queries read. Different code paths, sometimes different DBs. 2. Use when: read traffic is 5-10x writes, heavy search/joins, or you need independent scaling. 3. Common setup: write DB (normalized) → events/CDC

Sewajarnya nama brand aja Takumi

Junior to Senior engineer career progression isn’t about writing more code. 1. Junior: you ship tickets. 2. Mid: you own a feature end to end, tests, rollout, on-call. 3. Senior: you prevent incidents. You remove the next 10 tickets. You make others faster. Good signals I look

As a Golang Developer, Please switch jobs if you cannot clearly explain at least 10 of the following: Go memory model Happens-before (mutex, channels, atomic, WaitGroup) Data races and why “it works on my machine” is a lie Goroutine scheduling (G, M, P model) Preemption and

Abhishek Singh First rule: Never block the webhook on business logic. If the provider is sending 100 req/sec and I can only process 10, the webhook endpoint should do almost nothing: 1. Validate signature 2. Persist raw payload (durably) 3. Enqueue to a buffer (Kafka/SQS/Redis) 4. Return 200

ahh mungkin hanya perasaanku saja

In an interview they asked: “Design an API for flight booking.” Most people jump to endpoints. I didn’t. 1. Clarified scope: search vs reserve vs pay vs ticket, one-way vs multi-city, seats hold time (5 min? 15?), cancellations, refunds. 2. Defined entities: Flight, Fare,

Lucu

Facebook hit a classic cache stampede: one hot key expires, cache miss rate spikes, and suddenly 10k requests/sec dogpile the DB for the same row. Cache becomes a "thundering herd amplifier". This is how we prevent it: 1. Request coalescing (singleflight) Only 1 in-flight