Oolooroo
Articles about Oolooroo
Docker vs Kubernetes, Explained Like You’re New to Tech
Docker packs your app into a box. Kubernetes runs thousands of those boxes for you. Here's which one you actually need.
Your Load Balancer Isn’t the First Decision: DNS Routing Explained in 6 Policies
Simple, weighted, failover, latency, geo and IP-based routing, with the trade-offs behind each and when to use which.
How Blinkit and Zepto Actually Deliver in 10 Minutes - 10 Minutes Isn’t a Delivery Problem
Six architecture layers, 3,500+ dark stores and three ideas (proximity, prediction, parallelism) behind quick commerce
Caching in Distributed Systems: The 4 Patterns and 5 Traps Behind Every Fast System
Adding Redis takes an afternoon. Surviving a stampede, a hot key and stale data takes design. Here's the full guide.
Airbnb’s Architecture Evolution: 9 Bottlenecks That Forced a Monolith to Scale
Nobody designed it upfront. Here's the order each bottleneck hit, and the fix that followed, so you don't adopt step 9 on day one.
8 System Design Strategies That Separate Production-Ready Engineers From Everyone Else
Stateless services, caching, sharding, and five more patterns, what each one actually fixes, and why no single one is ever the whole answer
Millions of Prompts, One Second: The Architecture Behind ChatGPT
What happens between pressing Enter and watching the answer type itself out.
Behind Netflix’s Play Button
How local delivery, adaptive quality, and resilient systems keep video flowing worldwide
100 Billion Messages a Day: What Your Two Blue Ticks Are Really Telling You
The system design behind WhatsApp, explained through the grey and blue ticks you already know.
SQS vs SNS vs EventBridge vs Kinesis: Choosing the Right AWS Messaging Service
Four services, four fundamentally different delivery models, and the technical details that decide which one actually fits your system
How PostgreSQL Actually Processes a Query, From Connection to Commit
The internal mechanisms, shared buffers, WAL, MVCC, and the optimizer, that determine whether your database is fast or fragile
What Happens After You Scan a UPI QR Code?
Inside the architecture processing roughly 740 million payments every day
How JioHotstar Served 65.2 Million Live Viewers at Once
The seven-layer system design behind a world-record cricket stream, explained in under two minutes
Why Some Codebases Scale for Years While Others Collapse
Every developer has worked in a codebase they were afraid to touch. You change one line, and something breaks three services away. Other systems seem to absorb change easily: new features slot in, teams ship on their own schedules, and bugs stay contained. The difference usually isn’t talent. It’s a handful of design decisions that show up again and again in systems that last.
How a CDN Actually Makes the Internet Feel Instant
The architecture behind sub-second load times, and the decision framework for whether your system actually needs one
The Microservices Reference Architecture: A Layer-by-Layer Breakdown
What actually separates a production-ready microservices system from a distributed monolith wearing a microservices costume
Where Caching Actually Happens: 8 Layers From Browser to Database
Every layer in a modern request path caches something different, for a different reason — and picking the wrong layer to invalidate is how stale data reaches users
AI vs ML vs DL vs Generative AI: What Each One Actually Does
Four nested terms, four different training styles, four different outputs — and why conflating them leads to the wrong architecture decision
8 API Gateway Design Patterns and When to Use Each One
Eight architectures for eight different problems — and why the wrong one fails quietly, not loudly
The Firewall You Have Isn’t the Firewall You Need
Why "we have a firewall" is a single data point, not a security posture
How Companies Ship Code to Production in 8 Stages
Every day, teams ship code to production hundreds of times, and almost nothing breaks. It isn’t luck. It’s a loop of eight stages: plan, design, build, test, release, deploy, operate, and monitor and improve. I made a short explainer video that follows one small change through every stage, so you can see what actually happens between “I wrote some code” and “it’s live for real users.” No jargon required.
How RAG Works, Stage by Stage
LLMs don’t know your data. Ask one about your refund policy and it will hallucinate a fluent, confident, wrong answer, because its knowledge is frozen at training time and your private docs were never in it. RAG fixes this by changing the question from “what does the model remember?” to “what can we hand it right now?”