1. Cơ bản về cách blockchain thêm block Các blockchain như Ethereum, Bitcoin hoạt động như sau: Mỗi node trong mạng sẽ giữ một bản sao của blockchain hiện tại. Khi một miner (hoặc validator) tìm được block mới hợp lệ, nó sẽ broadcast block đó cho mạng. Các node nhận block mới sẽ xác minh, nếu hợp lệ thì thêm vào chain của mình. Vấn đề xảy ra khi nhiều miner có thể tìm được block mới gần như cùng một lúc. Giả sử mạng lưới đang dừng ở block thứ 99. Miner A tìm ra block 100A và broadcast nó. Ở một nơi khác, Miner B cũng tìm ra một block 100B và broadcast. Do độ trễ mạng, các node khác nhau có thể nhận và thêm vào chuỗi khác nhau: 100A hoặc 100B. ...

Xin chào các bạn, Trong bài viết này, tôi sẽ chia sẻ một số kinh nghiệm cá nhân khi phải lựa chọn giữa hai loại cơ sở dữ liệu phổ biến là SQL và NoSQL, cụ thể là PostgreSQL và MongoDB. Bài viết không đi sâu vào lý thuyết SQL hay NoSQL là gì, mà sẽ tập trung vào góc nhìn thực tế khi phải đưa ra lựa chọn trong một dự án cụ thể. ...

🚀 This is the beginning of my Hugo + PaperMod powered blog hosted at thanhhv.github.io. Stay tuned for more!

🔢 Common Algorithm Problems in Tech Interviews This wiki stores a collection of popular algorithm problems that often appear in backend or fullstack interviews. 1. Arrays & Strings Two Sum Longest Substring Without Repeating Characters Sliding Window Maximum 2. Linked Lists Reverse a Linked List Detect Cycle in a Linked List Merge Two Sorted Lists 3. Trees Binary Tree Inorder Traversal Lowest Common Ancestor Serialize and Deserialize Binary Tree 4. Dynamic Programming Fibonacci (Top-down & Bottom-up) Longest Increasing Subsequence 0/1 Knapsack 5. Graphs BFS & DFS Template Detect Cycles in Directed Graph Dijkstra’s Algorithm 6. Others Top K Frequent Elements LRU Cache Median of Two Sorted Arrays

📌 Common Backend Interview Questions Below are categorized questions that frequently appear in backend developer interviews. 1. System Design How would you design a URL shortener? How to design a rate limiter? What are the differences between monolith and microservices? What is a message queue and when would you use one? 2. Databases What is the difference between SQL and NoSQL? How does indexing work in PostgreSQL? What is ACID and why is it important? How do you scale a relational database? 3. API Design What’s the difference between REST and GraphQL? How do you handle versioning in APIs? What are idempotent HTTP methods? How to secure an API? 4. Caching & Performance What is the role of Redis in a web application? How would you prevent cache stampede? What is a CDN and when should you use one? 5. Deployment & DevOps What is a container and how does Docker work? What are health checks in a production service? How would you set up CI/CD for a microservices architecture?

🎯 Goal: Ensure the system is stable, scalable, and production-ready before launching. 🧱 I. CODE & FUNCTIONAL Code reviewed and approved by at least one senior developer Unit test coverage ≥ 80% Edge cases handled properly No N+1 queries (GraphQL/ORM optimized) No leftover console.log or debugger in code No secrets or tokens hardcoded 🚦 II. CI/CD & DEPLOYMENT CI pipeline set up (build → test → deploy) Rollback plan in place (script or quick revert strategy) Staging/UAT environment available and tested .env variables configured correctly (not copied from local) Proper branch/tag naming (e.g. release/v1.0.0) 🔒 III. SECURITY Auth applied on all sensitive routes (JWT, OAuth, API keys) Role & permission system tested No publicly exposed sensitive endpoints (/admin, /debug) Checked for XSS, SQLi, path traversal vulnerabilities Rate limiting/throttling in place for public APIs Security headers enabled (helmet, CORS, etc.) 📈 IV. PERFORMANCE & SCALING Load tested for target RPS (e.g. ≥ 10k req/s) Concurrent users simulated (e.g. 1k logins) DB indexed properly (EXPLAIN ANALYZE used) Redis, CDN, or in-memory caching applied File/media offloaded to external storage (e.g. S3) API only returns necessary data (REST/GraphQL select) 🧠 V. MONITORING & ALERTING Logging integrated (Winston, Pino, ELK, Datadog, etc.) Logs include traceId, userId, route context Metrics collection: Prometheus / Grafana / OpenTelemetry Alert configured for: High error rate (5xx) CPU > 80% Memory full Queue backlog increasing ⚙️ VI. INFRASTRUCTURE Load balancer configured and stable Auto-restart and healthcheck ready (PM2, Docker, k8s readiness) Database backup and restore tested Horizontal scaling/pod autoscaling set up if using k8s .env.production encrypted or managed via Vault 📋 VII. OTHER Changelog written (CHANGELOG.md) Release announcement sent to relevant teams Post-deploy checklist created (Smoke test UI/API) QA checklist completed (login, checkout, error, etc.) Graceful fallback UI for error handling /status, /health, /metrics, /version endpoints ready ✅ READY TO GO LIVE? Goal Ready? Code clean and stable ✅ Load test passed ✅ Monitoring in place ✅ Rollback plan confirmed ✅ Stakeholders notified ✅ 🔁 POST GO-LIVE ACTIONS Real-time system monitoring active User feedback monitored Latency/error rate observed 24h go-live report or postmortem (if needed)

Reading is not only for fun, but also to develop clear thinking and deeper perspective. Here’s a list of non-technical books that can help backend engineers grow in mindset and soft skills. 📘 Psychology & Thinking Thinking, Fast and Slow - Daniel Kahneman The One Thing - Gary W. Keller, Jay Papasan 🌍 Philosophy & Life Naval Ravikant: Để thịnh vượng và hạnh phúc - Eric Jorgenson … will be updated more ...