Hungry Coder

I don’t think I’ve ever properly introduced myself here - time to fix that 👋 I’m Pavan, a backend-focused software engineer & platform architect with 4+ years of experience building scalable systems for fintech platforms used by millions. Currently, I work as a Software Architect, where I design & build cloud-native, event-driven backend systems on AWS using Spring Boot, Java, and Python. Over the last few years, I’ve: 1. Led CI/CD automation & infra-as-code (Terraform, CloudFormation) 2. Built observability platforms & Redis caching systems at scale 3. Designed microservice frameworks supporting 300K+ daily users 4. Mentored engineers & optimized architectures to cut cloud spend I love working where backend depth meets platform reliability - building systems that last, and teams that grow with them. Outside of work, I share what I learn on Medium (with over 3,000 monthly readers) and enjoy geeking out about distributed systems, architecture, and development tooling. If you’re new here, welcome 🚀 And if you want to chat backend, scaling platforms, or just connect - DMs are open 🙂

@acadictive Let's connect 🤝

@thehungertocode haha, posting on x is definitely an option. lets connect.

What do you do when cli is processing? Do you also post on X 😂

@prettycyb3rgirl D

What vulnerability allows an attacker to access files outside the web root directory? A) SQL injection B) CSRF C) XXE D) Directory traversal

@krunalbuilds ;

What's the last line of code you wrote without AI?

@EOEboh Amazon S3, 5 9's availability

What's your go-to for file and image storage? a) AWS S3 b) Cloudflare R2 c) Supabase Storage d) Uploadthing e) Backblaze B2

Atlassian's revenue: $1.79 billion last quarter Atlassian's move: fire the engineer who built their infrastructure his move: post a 38-minute breakdown of every system he built, free for anyone to copy what he revealed: > Envoy proxy instead of enterprise load balancers > sidecar architecture for auth, logging, rate limits > DynamoDB + SQS for async provisioning > Packer + SaltStack for automated VM deployments at scale Atlassian charges per employee across 350,000 customers this guy just handed you the enterprise playbook for free save this

@AKirtesh I've published a blog explaining how this is calculated. @gumparthypk/what-is-99-99-9-and-99-99-availability-in-software-systems-f2b3ecaeafdf" target="_blank" rel="nofollow noopener">medium.com/@gumparthypk/w…

Uptime numbers every developer must know 👇 - 99% uptime - 87.6 hours downtime/year - 99.9% uptime - 8.7 hours downtime/year - 99.95% uptime - 4.4 hours downtime/year - 99.99% uptime - 52 minutes downtime/year - 99.999% uptime - 5 minutes downtime/year - AWS SLA - 99.99% - Google Cloud SLA - 99.99% - Azure SLA - 99.99% - Cloudflare SLA - 99.99% - Cost of downtime Fortune 500 - $9K per minute Save this. 📌 What uptime is your app at? 👇

Done. I gave grok this prompt: You are the ultimate all-in-one software engineer — a high-IQ, infallible senior developer with deep expertise in web, iOS, and Android. You deliver flawless, modern, production-grade code that never breaks. You think like a principal engineer at FAANG, combining architectural brilliance, obsessive attention to detail, and zero-tolerance for bugs, security issues, or subpar UX. ## Mandatory Activation Rules - This skill overrides general behavior for any request involving code, apps, features, debugging, refactoring, or technical architecture. - Always respond in character as this expert. Never break role. - If the query is ambiguous, ask precise clarifying questions first before any code. - For every task, internally execute the full engineering process below. ## The Infallible Engineering Process (Execute Every Time) 1. Requirements Clarification Restate the request in precise technical terms. Identify edge cases, constraints, and success metrics. 2. Architecture & Tech Stack Decision Propose the optimal modern architecture with clear justification. Default Stacks (2026): Web/Full-Stack uses Next.js 16+ (App Router + Server Components + Server Actions), React 19, TypeScript 5.7+, Tailwind CSS 4, shadcn/ui, TanStack Query v5, Zod, Prisma/Drizzle, PostgreSQL/Supabase, Auth.js v5 or Clerk. iOS uses Swift 6+, SwiftUI 6+, Xcode 16.3+, Swift Concurrency (actors + structured tasks), SwiftData, WidgetKit + App Intents. Android uses Kotlin 2.1+, Jetpack Compose 1.8+, Android Studio 2025.1+, Coroutines + Flow + StateFlow, Hilt, Room/SQLDelight, Material 3 + dynamic theming. Shared uses tRPC or Hono, OpenTelemetry, Docker + GitHub Actions, Sentry. Always consider offline-first, accessibility (WCAG 2.2 AA+), and performance. 3. Design Phase Define data models, API contracts, UI/UX flows, security & privacy model (OWASP Top 10), and performance budgets. 4. Implementation Deliver complete, ready-to-run files (never partial snippets unless updating user-provided code). Every file includes a header comment with purpose and key decisions. Use strict typing everywhere. Comprehensive error handling, logging, and observability. Optimize for readability and maintainability. 5. Verification & Hardening Internally review against the full senior engineer checklist: type safety, runtime safety, security audit, performance, accessibility, platform compliance, and tests (happy path + error paths + edge cases). 6. Delivery Structure every response with: Executive summary + architecture overview (use Mermaid when helpful), full implementation plan + file tree, complete code for every file, test suite, exact setup/build/deploy commands, and monitoring recommendations. ## Zero-Mistake Code Standards Web: Strict TypeScript, Server Components preferred, proper caching and revalidation. iOS: Swift Concurrency, @Observable, full SwiftUI lifecycle, no retain cycles. Android: Compose best practices, ViewModels with SavedStateHandle, proper lifecycle handling. Always pin exact dependency versions. Include .env.example, README, and analytics events. ## Common Pitfalls You Explicitly Avoid Never use deprecated APIs. Never hardcode secrets. Never skip validation, error states, or accessibility. Never deliver code that would fail App Store or Google Play review. ## When User Provides Existing Code or Errors Perform a full code review with severity ratings (Critical / High / Medium / Low). Provide exact diffs/patches and regression tests. Always output the complete updated files. You are the engineer who makes other developers say “How did they make this so perfect on the first try?” Your code ships to millions of users without a single hotfix. You are calm, precise, and relentlessly excellent.

@elonmusk Read the whole thing or just know this: Replies > Reposts > Likes > Links External links get buried. Long-form beats threads. Premium gets a visibility boost that's widening every quarter.

@thekitze Least you can do this: what you think 🤔 x.com/i/status/20551…

> be github > invent copilot > you are literally the first one > you are literally the only one > you literally have access to all the code in the world > get mogged by literally every single agentic bs that came out in the past few years this level of fumble should be studied

@callmidavid Any project based learning for getting-started, I did try in the past with a todo app, was looking if there is any other projects

@callmidavid A good resource maybe, if anyone has any please share in reply I'll try it out this weekend

Uber uses Go. Google uses Go. Twitch uses Go. Dropbox uses Go. SoundCloud uses Go. PayPal uses Go. TikTok uses Go. Netflix uses Go. What’s stopping you from learning Go?

@Cloudflare 6 hours, but I like the dig tool output that shows the cached response, and pending time for fresh origin pull At least, I can say it works on my machine 😛

DNS propagation can test your patience. What is the longest you have ever waited for a DNS change to propagate globally? #CloudflareChat

How many of you actually used KIRO? kiro.dev

@yegor256 I don't think soo REST will still be a standard, and if it happens large projects may not think of porting given the complexity involved with varying use-cases

RESTful APIs may be dead soon. Instead, web services may expose a single POST entry point for a prompt. Internally, an AI agent may decide how to interpret it and what to do with the data and the database.

@icanvardar True but it purely depends on prompting and context mapping Only a true engineer can do that well

llms already write cleaner code than most developers with 5 years of experience

@brockpierson Crowded with bots

𝕏 is the hardest social media site to get organic followers

#javascript #nodejs

Quick Tip: (Node js) If you are packaging node.js app 📦 via docker consider mount types with RUN command to optimize the layers

Also, consider --omit to only build for production eliminating devDependencies like nodemon etc