Blog
Writing on GPUs, LLMs, MLOps, Kubernetes — and mindset · 3525 posts
#2026-03 765#english 592#culture 264#deep-dive 254#kubernetes 249#career 229#ai 219#llm 209#devops 196#2026-04 146#security 143#database 114#observability 113#communication 109#history 107#architecture 100#productivity 96#finance 88#economy 84#mindset 81#psychology 80#ai-papers 79#food 78#it 78#travel 78#deep-learning 77#japanese 77#networking 77#performance 72#business-travel 70#linux 70#gpu 69#ai-agent 66#cs-fundamentals 63#postgresql 61#rag 58#self-improvement 55#learning 53#mlops 53#python 52
BGP Internet Routing Deep Dive 2025: AS, Peering, Path Selection, RPKI, Anycast — How the Internet Really Works
How does the internet appear as "one"? A complete walkthrough of BGP, Autonomous Systems, Peering vs Transit, Path Selection, RPKI, and route hijacking defense — the routing machinery running behind the scenes.
2026-04-15 · 14 min read #bgp#internet-routing#as#peering#rpkiCUDA GPU Programming Model Deep Dive — SIMT, Memory Hierarchy, Tensor Core, Kernel Optimization (2025)
The engine behind ChatGPT, Stable Diffusion, and Sora — NVIDIA GPUs and CUDA. This post dissects the CUDA programming model from scratch: GPU hardware architecture (SM, Warp, CUDA Core), SIMT vs SIMD, thread hierarchy (G
2026-04-15 · 15 min read #cuda#gpu#nvidia#ai#machine-learningArchitecture Case Studies — Netflix, Stripe, Cloudflare, Shopify in 2025
Netflix's 7,000 microservices, Stripe's Ruby monolith, Cloudflare's Workers ecosystem, Shopify's Black Friday scaling — real architecture drawn from public engineering blogs.
2026-04-15 · 13 min read #architecture#case-study#netflix#stripe#cloudflareCPU Cache & Memory Hierarchy Complete Guide 2025: L1/L2/L3, Cache Line, False Sharing, Performance Tuning
Everything about the memory hierarchy: L1/L2/L3 cache, cache line, MESI protocol, false sharing, cache-friendly data structures, NUMA, mechanical sympathy, and hands-on performance tuning.
2026-04-15 · 11 min read #cpu-cache#memory#performance#cache-line#false-sharingApache Spark Internals Deep Dive 2025: RDD, Catalyst Optimizer, Tungsten, Whole-Stage Codegen, Shuffle
Why Spark runs the same query 100x faster than MapReduce. RDD to DataFrame/Dataset, Catalyst optimizer, Tungsten project, whole-stage code generation, and shuffle optimization — a complete internal analysis.
2026-04-15 · 13 min read #spark#catalyst#tungsten#rdd#whole-stage-codegenAI-Native UI and Generative UX 2025 — Complete Guide: Streaming UI, Tool Use Rendering, AI SDK (Vercel), LangChain UI, LlamaIndex.TS UI, Conversational Interfaces, Feedback and Correction UX, Uncertainty Visualization, Human-in-the-loop, 2025 AI Product UX Benchmark — Season 6 Ep 3
While AI generates a response, what does the user see? The full scope of 2025 AI-Native UI. Streaming UI (token-by-token render), Tool Use to dynamic card rendering, Vercel AI SDK, LangChain UI, LlamaIndex.TS UI, UX prin
2026-04-15 · 12 min read #ai-ui#generative-ui#streaming-ui#tool-use#ai-sdkData Orchestration 2025 Complete Guide: dbt, SQLMesh, Dagster, Airflow, Prefect, Data Contracts, CI/CD (2025)
Season 5 Ep 4. If engines query, orchestrators govern the pipelines. The standardization of dbt and its limits, the revolt of SQLMesh, the asset-centric model of Dagster, the evolution of Airflow 2.x and 3.0, the rebirth
2026-04-15 · 13 min read #data-orchestration#dbt#sqlmesh#dagster#airflowModern Computer Architecture — CPU Pipelines, Out-of-Order, Caches, Branch Prediction, Meltdown, Apple Silicon, ARM, RISC-V, SIMD, GPU Deep Dive (2025)
Why arrays beat linked lists, why a branch can be 10x slower, how M1 beat Intel. CPU pipelines, Out-of-Order, L1 to L3 caches, branch prediction, Meltdown/Spectre, Apple Silicon, ARM/x86/RISC-V, SIMD, GPU SM/Warp, HBM/CX
2026-04-15 · 15 min read #computer-architecture#cpu#cache#branch-prediction#apple-siliconCompiler and Linker Internals — From #include to an Executable: ELF, Symbols, Static/Dynamic Linking, Relocation, LTO (2025) ♪ Listenable
Tracing the journey of "hello.c" becoming an executable, atom by atom. The four phases of preprocess / compile / assemble / link, ELF file layout, symbol tables and relocation, the archive structure of static libraries,
2026-04-15 · 20 min read #compiler#linker#elf#symbol#static-linkingCPython Bytecode Interpreter Deep Dive — ceval.c, Specializing Adaptive, PEP 659, Copy-and-Patch JIT (2025)
Python is slow, but most developers do not actually know why, or how it is getting faster. This post dissects CPython's heart, the ceval.c interpreter, from scratch: source to AST to bytecode, Code/Frame/Cell objects, th
2026-04-15 · 16 min read #python#cpython#bytecode#interpreter#jitClickHouse MergeTree — The Complete Guide to Columnar, Sparse Index, Part, and Materialized View (2025) ♪ Listenable
Everything about ClickHouse MergeTree — the physical layout of columnar storage, parts and granules and the sparse primary index, why ORDER BY matters so much, projections and skip indexes, real-time aggregation with mat
2026-04-15 · 23 min read #clickhouse#olap#columnar#mergetree#analyticsPerformance & Core Web Vitals 2025 — LCP, CLS, INP, RUM, Image/Font/JS Optimization, Edge, CDN, HTTP/3 (S6 E6)
Performance is not a feature — it is the precondition for every feature. The full 2025 Core Web Vitals map: LCP, CLS, INP (FID successor), RUM vs synthetic, critical rendering path, JS bundle diet, image/font optimizatio
2026-04-15 · 7 min read #performance#core-web-vitals#lcp#cls#inpConsistent Hashing Complete Guide 2025: Virtual Nodes, Jump Hash, Rendezvous, and the Core Building Block of Distributed Systems
How DynamoDB, Cassandra, and Memcached minimize data movement when nodes are added or removed. A 700-line deep dive on Consistent Hashing, from first principles to virtual nodes (vnodes), Jump Hash, Rendezvous Hashing, a
2026-04-15 · 16 min read #consistent-hashing#distributed-systems#sharding#virtual-nodes#jump-hashCDC (Change Data Capture) Complete Guide 2025: Debezium, Kafka, Data Sync, Outbox Pattern
Everything about CDC — how it works (WAL/binlog based), Debezium architecture, PostgreSQL Logical Replication, MySQL binlog, Outbox Pattern, data warehouse sync, Kafka integration, migration, common pitfalls.
2026-04-15 · 11 min read #cdc#debezium#kafka#postgresql#mysqlComplete Guide to CDN & Edge Caching Strategies 2025: Cache Invalidation, ETag, Stale-While-Revalidate
Everything about CDNs and caching. HTTP cache headers (Cache-Control/ETag/Vary), how CDNs work, edge caching, the hard problem of cache invalidation (one of the two hard problems), Stale-While-Revalidate, and a compariso
2026-04-15 · 14 min read #cdn#cache#edge-computing#http-cache#cache-invalidationCompilers and Modern Language Runtimes — LLVM, JIT, GC, V8 TurboFan/Maglev, Inline Caching, Escape Analysis, Rust Monomorphization Complete Guide (2025)
Everything that happens from a line of code to CPU execution. LLVM's dominance, V8's 4-tier JIT, Hidden Classes and Inline Caching, Escape Analysis, GC lineage from Mark & Sweep to ZGC, Go's work-stealing scheduler, Rust
2026-04-15 · 11 min read #compiler#runtime#llvm#jit#garbage-collectionBig Tech Engineering Culture Deep Dive: How Google, Meta, Amazon, Netflix, and Apple Differ (2025)
Google's Design Doc, Meta's Move Fast, Amazon's 6-page Memo, Netflix's Keeper Test, Apple's secrecy. A comparison of the actual engineering cultures at the five Big Tech companies and Korean Big Tech (Naver, Kakao, Coupa
2026-04-15 · 16 min read #culture#engineering-management#google#meta#amazonModern Code Review and Merge Pipelines — PR, Merge Queue, Stacked PRs, Monorepo, AI Review, Trunk-Based, Husky, Semgrep Deep Dive (2025)
The daily activity we talk about least. PR sociology, Code Owners, Merge Queue (GitHub/Mergify/Aviator/Graphite), Stacked PRs (Graphite, Sapling, Jujutsu), Monorepo vs Polyrepo 2025, Nx/Turborepo/Moon/Bazel, AI review (C
2026-04-15 · 12 min read #code-review#pull-request#merge-queue#monorepo#stacked-prsChaos Engineering Deep Dive — Netflix Simian Army, LitmusChaos/Chaos Mesh, AWS FIS, Game Day
Why Netflix started randomly killing production servers in 2010. From Chaos Monkey philosophy and the 4 principles, to the full Simian Army, LitmusChaos/Chaos Mesh/AWS FIS comparison, Game Day design, and blameless postm
2026-04-15 · 10 min read #chaos-engineering#sre#reliability#netflix#kubernetesComplete Guide to Async I/O Models 2025: epoll, io_uring, Reactor/Proactor, and async/await Internals
Everything about async I/O. Blocking vs non-blocking vs async, the select/poll/epoll/kqueue evolution, the iouring revolution, Reactor/Proactor patterns, Tokio/libuv internals, and what async/await actually does.
2026-04-15 · 12 min read #async-io#epoll#io-uring#reactor#proactor