Blog
Writing on GPUs, LLMs, MLOps, Kubernetes — and mindset · 3517 posts
#2026-03 765#english 592#culture 264#deep-dive 254#kubernetes 247#career 229#ai 216#llm 208#devops 193#2026-04 146#security 141#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 60#rag 58#self-improvement 55#learning 53#mlops 53#ai-platform 51
What Java Value Objects (JEP 401) Change — What You Gain and Lose by Giving Up Identity
At 00:45 UTC on July 31, 2026, the JEP 401 Value Objects implementation merged into the OpenJDK mainline. Bundled into a single commit with JEP 539 (Strict Field Initialization), it added roughly 208,000 lines across 1,8
2026-07-31 · 16 min read #java#jvm#valhalla#jep-401#performanceHow to Read the 2026 Data Tooling Landscape — Layer-by-Layer Decisions, Not a Tool List
On July 28, 2026, "A Data Tooling Landscape Guide for Developers" landed on GeekNews and picked up 51 points. It is a good map, but a map does not choose your destination for you. This post redraws the same landscape acr
2026-07-31 · 16 min read #data-engineering#lakehouse#iceberg#duckdb#data-platformHow Engineers Should Read a Go-to-Market Playbook — Where Distribution Beats Product Quality
In late July 2026, an NFX post titled "Upgraded Go-to-Market Playbook" showed up on GeekNews. It is a perspective piece written by a VC, and the numbers inside are curated anecdotes, not a sample. This post translates th
2026-07-31 · 14 min read #startup#go-to-market#product-led-growth#developer-tools#careerRunning AI Agents in Production — Idempotency, Budgets, and Confidently Wrong Answers
Moving an agent from prototype to production exposes an operational surface you only discover late: idempotency for retried tool calls, budgets and step limits, observability over non-deterministic control flow, per-tool
2026-07-31 · 16 min read #ai#agents#observability#reliability#mcpGCC's AI Policy and Open Source's Choice — What Is the 15-Line Boundary Actually Protecting
On July 29, 2026, the GCC steering committee adopted the recommendations of its AI Policy Working Group. The core rule is that legally significant contributions containing or derived from LLM-generated content will not b
2026-07-31 · 12 min read #ai#open-source#gcc#licensing#governanceA Practical Procedure for Moving a Codebase with AI — Stand Up the Judge First, and Measure Review Rate Instead of Lines
This post distills the procedure common to the large-scale LLM migrations published in the first half of 2026 (535,000 lines of Zig to Rust, 165,000 lines of Python to TypeScript) into an executable order of operations.
2026-07-31 · 13 min read #ai#migration#refactoring#testing#engineeringWhat DeepSeek V4 Flash Actually Changes — A Cost-per-Capability Question, Not a Leaderboard One
On July 31, 2026, DeepSeek moved its V4-Flash API into public beta. The architecture is identical to April's preview — 284B total parameters, 13B active MoE, 1M context — and the only thing that changed is post-training.
2026-07-31 · 11 min read #ai#llm#deepseek#inference-cost#moeBuilding an Internal Knowledge Base on an LLM — Permission-Aware Retrieval, Freshness, and the Eval Set You Need Before Launch
The account Cerebras published on July 15, 2026 of how they built their internal knowledge base exposes the internals of a system fielding more than 15,000 queries a day from humans, automation, and agents alike. But the
2026-07-31 · 14 min read #ai#rag#enterprise-search#llm#platform-engineeringHow AI Found Two Years' Worth of Chrome Bugs in a Single Month — Reading the Number 1,072 Correctly
According to an announcement Google made on July 30, 2026, Chrome 149 and 150 — the two versions released in June alone — fixed 1,072 security bugs, more than the 1,036 fixed across the previous two years and 23 mileston
2026-07-31 · 12 min read #ai#security#chrome#fuzzing#vulnerabilityThe Economics of Refactoring: When Does It Pay Off — Calculating With Change Frequency
The Economic Benefit of Refactoring, published on martinfowler.com on July 30, 2026, refactored a 17,000-line module in 15 steps and measured input tokens for the same repeated change request dropping from 159,564 to 27,
2026-07-31 · 15 min read #refactoring#engineering#technical-debt#ai#metricsWhat It Means to Generate 3D CAD From a Prompt — Mesh vs. B-rep, and the Constraint Bottleneck
A tool called CAID that generates 3D CAD models from a prompt showed up on GeekNews Show GN. The limitations its own creator disclosed summarize the state of this entire field with precision — no automatic dimension veri
2026-07-31 · 12 min read #ai#cad#llm#manufacturing#geometryEleven Days of Bun's Zig-to-Rust Rewrite — What Actually Transfers from a Large AI Migration
Follows Bun's 11-day move of 535,000 lines of Zig to Rust through the primary sources. From May 3 to May 14, 2026, it took up to 64 Claude instances, about 50 workflows, 6,502 commits, and roughly 165,000 dollars at API
2026-07-31 · 13 min read #ai#bun#rust#migration#testingThe Complete Guide to Installing k3s in an Air-Gapped Network — From Image Tarball Transfer to a Private Registry and Agent Joins
Covers the whole process of standing up a k3s cluster on a network with no internet path at all, command by command. It walks through downloading the airgap image tarball and the k3s binary on a connected machine, buildi
2026-07-31 · 20 min read #kubernetes#k3s#air-gap#on-premise#containerdAir-Gapped Kubernetes Day 2 Operations — How to Stop a Cluster From Dying of Certificate Expiry
What actually kills an air-gapped cluster is usually not some spectacular outage but certificate expiry. Let a year go by with nobody touching it and the API server quietly stops, and with no internet you cannot search y
2026-07-31 · 20 min read #kubernetes#air-gap#day2-operations#etcd#certificateDesigning an Air-Gapped Image Import Pipeline — skopeo, Harbor, and a Reimport Runbook That Doesn't Rot
This post designs the pipeline for a problem almost nobody in air-gapped Kubernetes actually gets right: repeatedly and safely bringing container images in. It covers managing the import list as code, the skopeo sync com
2026-07-31 · 17 min read #kubernetes#air-gap#harbor#skopeo#supply-chainAuthentication, Authorization, and IDOR — Why the Login Is Fine but You Can See Data That Belongs to Someone Else
Most real data breaches come from authorization, not authentication. The classic case is IDOR, where the login passes normally but changing only the ID in the URL returns another user resource verbatim. This post separat
2026-07-26 · 15 min read #security#authorization#idor#api#multi-tenantHow the Stanford Prison Experiment Fell Apart — What the Archives Revealed
The 1971 prison experiment run by Philip Zimbardo entered textbooks and feature films as the story of ordinary students who became brutal guards in six days. Yet the original data appeared not in a top-tier journal but i
2026-07-26 · 9 min read #psychology#paper-review#social-psychology#replication#research-ethicsHow Characters Change — The Grammar of Character Arc Design
Plot is the order of events; the arc is the change in a person. In a good story the two share the same scenes and push each other forward. Here are the three arcs — the positive one that lets go of a false belief, the ne
2026-07-26 · 12 min read #storytelling#character#fiction#narrative#craftThe Experiment Where One Dollar Beat Twenty — Cognitive Dissonance and Sixty Years of Rebuttal
Participants were made to do a deliberately boring task, then asked to lie about it. One group was paid one dollar, the other twenty dollars. The people who later rated the task as more enjoyable were the ones who had be
2026-07-26 · 13 min read #psychology#paper-review#social-psychology#decision-making#mindsetThe Milgram Obedience Experiment Reexamined — the Other Side of That 65 Percent
The Milgram experiment, begun at Yale in 1961, has been summarized in one sentence for decades: 65 percent of participants went all the way to the maximum voltage. But that figure is the value from one of some 24 variati
2026-07-26 · 11 min read #psychology#paper-review#social-psychology#obedience#research-ethics