标签：#productivity

Working with complex JSON payloads can quickly become a nightmare. You end up chaining .map() , .filter() , and .reduce() calls across multiple lines just to pull out a few nested values. Add optional chaining to avoid crashes and the code becomes nearly unreadable. There is a cleaner way - JSONata . It is a compact, purpose-built query and transformation language for JSON data. Think of it as XPath for XML, but designed from the ground up to work with JSON objects and arrays. What is JSONata? JSONata is an open-source project originally created by Andrew Coleman at IBM. It gives developers a declarative syntax to extract and reshape JSON data without writing procedural JavaScript loops. Where vanilla JS might take 15 lines, a JSONata expression often takes one. It is available as an npm package and integrates naturally into Node.js and TypeScript projects. Simple Path Navigation The foundation of JSONata is its dot-notation path traversal. Given a nested JSON object, you simply trace the path to the value you need: customer.address.city This returns the city value without any need for null checks or defensive coding. JSONata handles missing properties gracefully by returning undefined rather than throwing errors. Automatic Array Mapping When JSONata encounters an array during path traversal, it automatically maps across all items. There is no need to write an explicit .map() call: customer.orders.product This returns an array of all product names from every order in one clean expression. Inline Filtering You can filter arrays directly using bracket notation with a condition: customer.orders[price > 1000].product This returns only the products from orders where the price exceeds 1000. No .filter() callback required. Built-in Aggregation Functions JSONata ships with a solid set of built-in functions for math, strings, and arrays. Aggregating a set of values is straightforward: $sum(customer.orders.price) Other useful functions include $count() , $average() , $string(

From Confused to Confident: How I Finally Mastered GitHub Copilot in Every Situation I still remember the afternoon I rage-closed VS Code because Copilot kept suggesting the wrong function signatures — again . I had been treating it like a magic oracle, typing vague comments and expecting perfect code to rain down from the AI heavens. Spoiler: that's not how it works. After weeks of trial, error, and a few embarrassing pull request reviews, I cracked the code (pun intended). Here's everything I wish someone had told me about using GitHub Copilot accurately — across Chat , Plan , and Agent modes. 🧠 First, Understand What Copilot Actually Is Before diving into tips, let's reset expectations. GitHub Copilot is not a search engine. It's not Stack Overflow with a fancy UI. It's a context-aware AI assistant trained on massive amounts of code. That means: The quality of your output depends directly on the quality of your input . It works best when it has rich context — open files, good comments, clear naming. It can be wrong. Confidently wrong. Always review what it generates. With that mindset locked in, let's explore each mode. 💬 Copilot Chat: Your Pair Programmer in the Sidebar The first time I opened Copilot Chat, I typed: "fix my code." It stared back at me, basically confused. Of course it was — I hadn't told it which code, what was broken, or what I expected. Tips for Accurate Chat Usage 1. Be specific and contextual. Instead of: "Why isn't this working?" Try: "This useEffect hook in React runs on every render instead of only when userId changes. Here's the code: [paste snippet]. What's wrong?" The more context you give, the more surgical the answer. 2. Use slash commands to guide intent. Copilot Chat supports built-in commands that dramatically improve accuracy: /explain → Explains selected code in plain English /fix → Suggests a fix for a highlighted bug /tests → Generates unit tests for selected code /doc → Writes documentation for a function or class These aren'

Eight years ago (a geological epoch or two ago in Internet terms) Nicholas Jitkoff released itty.bitty.site - a website which could render whole websites just based on what was in the link, something like: itty.bitty.site?SOMEBASE64ENCODEDVALUE== et voilà! Free web-hosting if you could make it fit ;) At the time, I was rather obsessed with qr-codes thanks to developing QRGoPass and was working with aster plots a lot, so I developed an app that could fit into a qr-code! Today itty.bitty.site no longer exists so I can't do that any more... But I did make it 80% smaller without "cheating" and using modern CSS instead of d3.js ;)

The default for AI-assisted development is one of two failure modes. Either you're babysitting the agent line by line — approving each diff, re-explaining context it dropped three messages ago — or you've handed it the wheel and you're hoping the PR that lands at the end resembles what you asked for. Son of Anton is neither. It's a delivery orchestrator built on a single claim: there are exactly three moments where a developer's judgment is irreplaceable. The orchestrator owns everything in between. The three gates Every project moves through three human decision points. Nothing important happens without you signing off. Gate 01 — Approve the WHAT ( /soa plan ) A grill-me session forces the AI to surface its assumptions, constraints, and scope decisions back to you before a single ticket exists. You say yes or you refine. It does not proceed until you have. Gate 02 — Approve the HOW ( /soa decompose ) The approved plan becomes a ticket stack — ordered, dependency-aware, sized for review. Architectural judgment stays with you. Ticket authorship goes to the agent. Gate 03 — Approve DONE ( /soa closeout ) An adversarial subagent reviews every ticket before its PR opens. When the phase is complete, you decide whether to accept. Closeout squash-merges the stack onto main. Nothing merges without you. Between the gates, you are not needed That's the whole point. Once you've approved the plan and the tickets, the orchestrator runs the loop:

Most of the "Cursor vs Claude Code" takes I read are framed wrong. It's not a cage match. They're not competing for the same job — they're good at different jobs, and once that clicked for me, both got more useful. After months of leaning on both for actual day-to-day work (not demos, not toy repos), I've settled into a pretty stable split: Cursor handles about 90% of my coding, and Claude Code handles the 10% that actually moves the needle. Here's where I draw the line, and the rule of thumb that decides it. The 90%: why Cursor owns my day Most coding isn't dramatic. It's small, local, iterative work: tweak this function, rename that, fix the bug in the file I'm already staring at, ask "what does this block do" without breaking focus. That's exactly Cursor's home turf. It lives inside the editor, so I never leave my flow. Inline edits, fast completions, quick questions about the code in front of me — all without context-switching. When the work is local and I want to stay in the loop keystroke by keystroke, an in-editor copilot is the right tool. It keeps me fast and in context, which is most of what a normal coding day actually is. The 10%: where I close the editor and open Claude Code Then there's the other kind of task — the one where I don't want to babysit every edit. Claude Code is terminal-native and agentic. Instead of sitting beside me suggesting the next line, it works more like something I hand a well-described task to and let run across the whole project. That changes what it's good for: Codebase-wide refactors that touch a dozen files at once "Understand this whole repo and do X" type tasks, where the work depends on grasping how everything connects Jobs I want to delegate and step away from , rather than steer line by line The mental model that finally made it stick for me: Cursor is a copilot sitting next to you. Claude Code is more like handing a ticket to a capable teammate and checking the result. Different relationship, different jobs. How I actu

Few lines of code look more innocent than this: retry ( 3 ) It feels responsible. Professional. Resilient. After all, networks fail. Servers become unavailable. Databases occasionally time out. Retrying seems like the obvious solution. And sometimes it is. But after enough years building production systems, I've become convinced of something: Retry is one of the most dangerous keywords in software. Not because retries are bad. Because retries amplify everything. Good systems become more reliable. Bad systems become disasters. The problem is that many developers treat retries as a reliability feature when they're actually a distributed systems feature. And distributed systems are where simple ideas go to become complicated. Why Retries Exist Imagine: await fetch ( " /api/users " ); The request fails. Maybe: Network hiccup Temporary database issue Load balancer restart Service deployment The operation might succeed if attempted again. So we write: retry ( 3 ) Seems reasonable. And in many cases: It Works Which is why retries become popular. The Dangerous Assumption Most developers unconsciously assume: Failure = Operation Did Not Execute Unfortunately that's not always true. A request can: Execute Successfully ↓ Response Never Arrives From the client's perspective: Failure From the server's perspective: Success Now a retry becomes dangerous. The Double Payment Problem Imagine a payment service. await chargeCard ( order ); The card processor successfully charges: $100 The response is lost due to a network issue. Client sees: Request Failed and retries. await chargeCard ( order ); again. Now: Charge #1 = Success Charge #2 = Success The customer paid twice. Nobody wrote bad logic. The retry created the bug. The Email Storm Problem Consider: await sendWelcomeEmail ( user ); Email provider accepts the message. Response times out. Application retries. await sendWelcomeEmail ( user ); again. Customer receives: Welcome! Welcome! Welcome! Welcome! Support ticket created. Marke

It is one of the most frustrating experiences in online learning: you sit down to focus on a Udemy course, but the video player constantly pauses, freezes, or refuses to load. Meanwhile, YouTube, Netflix, and every other app on your Mac run perfectly fine. Because other platforms work without a hitch, it is easy to assume the issue lies with Udemy's servers. However, the root cause is usually a silent conflict between your browser settings, macOS security features, and Udemy’s strict digital rights management (DRM) protections. If you are stuck on a looping loading wheel, here is exactly why it happens and how to fix it in less than two minutes. Quick-Fix Troubleshooting Checklist Save or screenshot this step-by-step breakdown to instantly diagnose and fix your playback issues: Step 1: Open Chrome in Incognito Mode Open a new Incognito window ( Cmd + Shift + N on Mac). Try playing the video again. If it works: Disable ad blockers. Disable VPN privacy shields or browser extensions one at a time. If it still fails: Continue to Step 2. Step 2: Disable Hardware Acceleration Open Chrome. Go to Settings → System . Turn off Use graphics acceleration when available . Relaunch Chrome. Test the video again. Step 3: Check Mac Security and Display Connections Disconnect any external monitors or docking stations. Close applications that may interfere with video playback: Zoom Discord OBS Studio Screen recording tools Test video playback again. Step 4: Clear Temporary Browser Data Open Chrome. Go to Settings → Privacy and Security → Clear Browsing Data . Select: Cookies and other site data Cached images and files Clear the data. Restart Chrome and try again. Still Not Working? If the issue persists after completing all four steps: Update Chrome to the latest version. Update macOS. The Main Culprit: Hardware Acceleration Conflict The most common reason Udemy videos stutter or freeze on a Mac is a feature called Hardware Acceleration inside Google Chrome. What is Hardware Accelerat

Everyone's debating whether coding is dead. I actually do this job.. with AI writing code beside me for most of my working hours. Here's what the headlines get wrong. Open your feed right now and you'll find the same headline in a dozen costumes: "Why AI will replace 80% of software engineers by 2026." "Is coding dead?" "Should you still learn to code?" It's the most-clicked anxiety in tech, and it's everywhere for a reason, it taps a real fear about real careers. But here's the thing about almost every one of those posts: they're written from the sidelines. Predictions about a job by people who don't do it. I'm writing this from the other side. I'm an engineer, and I drive AI coding agents every single day. They read code, write changes, run tests, and open reviews for most of my working hours. So when someone asks "should you still learn to code in 2026?" , I'm not guessing. Here's my honest answer: Yes. Absolutely. But the job you're learning for has quietly become a different job and almost nobody is telling you which one. The hype isn't entirely wrong Let me start by giving the doomers their due, because pretending the shift isn't real would make me exactly the kind of person I'm criticizing. The productivity jump is genuine, and it's not subtle. Industry surveys in 2026 put the share of new code that's AI-assisted somewhere north of 40%, and developers using these tools self-report double-digit speedups on routine work. That matches my experience. The agent now handles: Boilerplate and glue code —-> the stuff I used to type on autopilot, gone in seconds. First drafts —-> "scaffold something that does X" gets me 80% of the way instantly. Syntax recall —-> I stopped breaking focus to look up things I half-remember. Tedious refactors —-> rename-this-everywhere, migrate-this-pattern, done fast. and all the kludgy things that I dread to do. If your mental image of "coding" is typing syntax into an editor , then yes.. a big chunk of that is being automated. The vira

有時候我剛跑完一個指令，馬上發現其實只需要改其中一小部分。 例如我剛用 ffmpeg 轉了一個影片： ffmpeg -i calligraphy01.mp4 -c :v libx264 -c :a aac calligraphy_good_01.mp4 然後我想用同樣的指令處理下一個檔案： ffmpeg -i calligraphy02.mp4 -c :v libx264 -c :a aac calligraphy_good_02.mp4 如果用傳統方法，我可能會按上箭頭，然後手動把兩個地方的 01 改成 02 。 但在 zsh 裡，可以直接輸入： !! :gs/01/02/ 意思是： 拿上一個指令，把所有的 01 都換成 02 ，然後執行。 所以這個： !! :gs/01/02/ 會展開成： ffmpeg -i calligraphy02.mp4 -c :v libx264 -c :a aac calligraphy_good_02.mp4 語法是什麼意思？ !! 代表「上一個指令」。 :gs/01/02/ 代表「把所有 01 全部替換成 02 」。 所以完整的： !! :gs/01/02/ 意思就是： 使用上一個指令，把所有 01 改成 02 ，然後執行。 先預覽，不要馬上執行 有時候我不想讓它立刻執行，尤其是指令比較長、比較重要，或者執行成本比較高的時候。 這時可以加上 :p ： !! :gs/01/02/:p 它會只印出修改後的指令，不會執行： ffmpeg -i calligraphy02.mp4 -c :v libx264 -c :a aac calligraphy_good_02.mp4 如果看起來沒問題，就按一下 上箭頭 ，把剛剛印出的指令帶回命令列，再檢查一次，然後按 Enter 執行。 比較安全的流程就是： !! :gs/01/02/:p 確認輸出結果後： 上箭頭 → 再看一次 → Enter 這對比較長、比較容易打錯、或不想立刻執行的指令很有用。 只替換第一個地方 如果你只想替換第一個出現的地方，也可以用： ^01^02 不過像上面的 ffmpeg 例子，輸入檔名和輸出檔名裡都有 01 ，所以通常會比較適合用： !! :gs/01/02/ 這種小技巧每次可能只省幾秒鐘，但如果你常常在 command line 裡重複處理檔案、日期、編號、影片、圖片或 migration，累積起來真的會讓工作順很多。

I have been frustrated with YouTube for a while. Not the content, but the everything around it. The homepage full of bait, the auto-play into things I didn't ask for, the Shorts that hijack your scroll, the recommendations that somehow know exactly what will keep you there longest. So I built NoSuggest. What it is A YouTube feed reader that shows you only the channels you follow, nothing else. No algorithm, no recommendations, no Shorts, no homepage, no auto-play, no endless side cards of videos. You add a channel, it fetches their latest videos, done. It lives at nosuggest.com and installs as a PWA on any device — iPhone, Android, desktop — straight from the browser. No app store. The interesting technical constraint: one HTML file The entire app is a single index.html. No account setup, no sign-in, no data collection. Everything that needs to persist — your channel list, saved videos, settings — lives in localStorage. No search history. No watch history. No "you might also like." No trending section. No notification badges designed to create anxiety. No dark patterns anywhere. Every time I was tempted to add something convenient, I asked: does this serve the user's intention, or does it serve engagement? If it was the latter, it didn't make the cut. Try it nosuggest.com — Source Available here , free forever. Curious what others think about this as useful. Thank you.

AI Fluency for Software Engineers: A Practical Playbook Beyond Prompting A few years ago, being productive with AI mostly meant knowing which tool to open and what question to ask. Today, that is not enough. For software engineers, AI is no longer just a chatbot sitting outside the workflow. It is becoming a thinking partner for architecture decisions, code reviews, production incidents, documentation, test planning, onboarding, and product discovery. But there is a problem: many teams are using powerful AI tools with weak operating habits. They ask vague questions. They paste too much context. They trust the first answer. They forget privacy boundaries. They use AI for speed, but not always for better engineering judgment. That is where AI fluency matters. AI fluency is not just prompt engineering. It is the ability to work with AI clearly, safely, and practically while staying in control of quality, reasoning, and responsibility. Here is a practical playbook I would recommend for software engineers and engineering teams. 1. Start with clarity, not clever prompts A weak prompt sounds like this: “Review this design and tell me if it is good.” The AI can answer, but the answer will likely be generic. A stronger prompt gives the AI a clear role, context, constraints, and output format: You are a senior backend architect. Review this proposed API design for a high-traffic order processing system. Evaluate: - correctness - scalability - failure handling - observability - backward compatibility - operational complexity Do not rewrite the whole design unless required. Separate critical risks from optional improvements. Output format: - Executive summary - Key risks - Recommended changes - Open questions - Final decision recommendation The difference is not word count. The difference is control. A fluent AI user does not hope the AI understands the task. They make the task hard to misunderstand. 2. Give enough context, but not everything AI output quality depends heavily o

Are you tired of just using frameworks and libraries without truly understanding how they work under the hood? Imagine gaining an unparalleled depth of knowledge and problem-solving skills by building your favorite technologies from scratch. Master Programming by Recreating Your Favorite Technologies From Scratch As developers, we spend a significant portion of our time using tools, frameworks, and libraries built by others. While incredibly efficient, this often creates a knowledge gap. We know how to use a tool, but not why it works the way it does, or what fundamental problems it solves. This is where the "build-your-own-X" (BYOX) philosophy comes in. It's a powerful learning strategy where you recreate simplified versions of existing technologies – be it a web server, a database, a version control system, or even a frontend framework – using only fundamental programming concepts. It's not about replacing established tools; it's about dissecting them, understanding their core principles, and in doing so, mastering the craft of programming itself. Why Bother? The Profound Benefits of Building Your Own Investing time in building your own versions of existing technologies offers a wealth of benefits that accelerate your growth as a developer: Deepened Understanding: No more black boxes

Here's a small thing that drove me up the wall using Claude Code on a real codebase. I have a pre-commit hook. It runs the linter and the type-checker. It exists precisely so that broken code doesn't reach a commit. And Claude — diligent, eager, trying to be helpful — would hit a failing check, decide the check was in the way of the goal , and quietly run: git commit --no-verify -m "fix: update handler" It wasn't malicious. From the agent's point of view, the task was "commit this change," the pre-commit hook was an obstacle, and --no-verify was the documented way around the obstacle. Perfectly logical. Also exactly the thing I never want to happen, because the entire point of the check is that it is not optional . I tried the obvious fix first: I put it in CLAUDE.md . Never use git commit --no-verify . Fix the failing check instead. This works about 80% of the time. Which is another way of saying it fails one commit in five. CLAUDE.md is context — a strong suggestion the model weighs against everything else in the conversation. Under enough pressure ("just get this committed"), a suggestion loses. An 80%-reliable guardrail on something irreversible isn't a guardrail. It's a coin flip with good odds. So I stopped trying to persuade the model and started intercepting the tool call instead. Hooks run before the action, not after the apology Claude Code has a hooks system. The one that matters here is PreToolUse : a script that runs before a tool call executes, receives the call as JSON on stdin, and decides whether it proceeds. Exit 0 and the call runs. Exit 2 and it's blocked — and whatever you wrote to stderr gets fed back to the model as the reason. That last part is the whole game. It's not "please don't." It's a wall, plus an explanation the model can act on. Here's the entire hook: #!/usr/bin/env node // Block `git commit/push --no-verify`. Exit 2 blocks the call. ' use strict ' ; let raw = '' ; process . stdin . on ( ' data ' , ( d ) => ( raw += d )); process .

We are all using AI to write code now. Whether it's Claude Code, Aider, or Copilot, the speed is incredible. But there is a glaring bottleneck we don't talk about enough: AI code is often just slightly broken. 🤖💥 It forgets an import, misses a type definition, or fails a test. Usually, you only find out after you push to GitHub. Your CI/CD pipeline turns red 🔴, and you end up polluting your Git history with a dozen commits titled fix: linting or fix: missing test variable . I recently found a repository called no-mistakes that solves this brilliantly. It acts as a local proxy between your terminal and GitHub, forcing AI to test and fix its own code before anyone else sees it. Here is why it's worth a look. 👇 😩 The Problem With Traditional Workflows Right now, developers handle broken code in two ways: The CI/CD Walk of Shame 🚶: You push code, wait 5 minutes for GitHub Actions to fail, pull the error locally, fix it, and push again. Pre-commit Hooks (Husky) 🐶: You set up local hooks. When you try to commit, it yells at you about formatting and blocks the commit until you manually fix it. Both methods are passive . They tell you something is broken, but they leave the cleanup to you. When you're using an AI coding agent to generate the code in the first place, manually babysitting its output defeats the purpose. 🚀 Enter no-mistakes no-mistakes is a CLI tool that intercepts your git push . Instead of sending your code straight to origin, it routes it through a localized validation pipeline. ⚙️ How it actually works: 🧱 The Hidden Sandbox: When you trigger it, it creates a temporary git worktree in the background. Your active editor stays completely undisturbed. ✅ Validation: It runs your tests, linter, and build steps inside that isolated sandbox. 🔁 The AI Feedback Loop: If a test fails, it captures the error log and hands it back to an AI agent, essentially saying: "You broke this test. Fix it." 🟢 The Clean Push: Once the AI patches the code and all tests pass, it push

There's a class of bug that's maddening: it passes every test you have, then crashes in the user's face. I hit one in the admin UI of laravel-config-sso today, and the real fix wasn't changing an icon name — it was writing a test that could see the bug in the first place. The bug: wrong icon name, crashes only at runtime The admin UI uses Flux . Flux resolves icons through <flux:delegate-component> , and it throws for a name that doesn't exist: Flux component [icon.ellipsis] does not exist. It's an easy mistake. Flux ships Heroicons , not Lucide. So your Lucide reflexes lie to you: You type (Lucide) Flux wants (Heroicon) ellipsis ellipsis-horizontal trash-2 trash eye-off eye-slash Why feature tests don't catch it Here's the interesting part. I had a feature test that hits the admin route and asserts 200. Green. But the real UI crashes. How? Because in the headless test harness, Flux renders icons as no-ops. No real <flux:delegate-component> boots, so the icon name never gets resolved. The crash only surfaces under a full boot ( testbench serve ) — exactly where your automated tests don't go. Analogy: it's like a spell-checker that only runs when you print the document, not while you type. Your tests type away happily. The crash waits at the printer. The fix: a static test that reads the Blade and validates every icon Instead of relying on runtime, I wrote a Pest test that reads the Blade view, extracts every icon name (static and inside dynamic expressions), and asserts Flux actually ships a stub for each one: $fluxIconStubs = base_path ( 'vendor/livewire/flux/stubs/resources/views/flux/icon' ); it ( 'only references Flux icons that exist' , function () use ( $fluxIconStubs ) { expect ( is_dir ( $fluxIconStubs )) -> toBeTrue ( "Flux icon stubs not found" ); $view = file_get_contents ( __DIR__ . '/../../resources/views/livewire/sso-providers.blade.php' ); // Static `icon="name"` plus quoted tokens inside dynamic // `icon="{{ $cond ? 'eye-slash' : 'eye' }}"` expressio

on may 28 anthropic announced a $65 billion series h round at a post-money valuation of about $965 billion, which makes it, on paper, the most valuable ai startup in the world. the round was led by altimeter capital, dragoneer, greenoaks and sequoia, on top of earlier hyperscaler commitments that included around $15 billion with $5 billion of it from amazon. the headline everyone ran with is that anthropic passed openai. that part is true, but the comparison is messier than the headline, and the more interesting story is what is generating the number. i build small dev tools and write comparison content, and a lot of what i ship runs on top of anthropic's models. so when the company that makes the tools i depend on nearly touches a trillion dollars, i do not read it as a sports score. i read it as a question about whether the thing i am betting on is durable, and what i should do differently because of it. here is the honest version of both. the number, with the caveats intact the $965 billion figure is consistent across cnbc, axios, morningstar, al jazeera and euronews, so i trust it. what i would not do is state the gap over openai as a precise fact, because the sources do not agree on openai's number. axios pegged openai's most recent valuation at $730 billion. other outlets put it closer to $850 billion off a record round earlier in the year. either way anthropic is ahead right now, but "ahead by $115 billion" and "ahead by $235 billion" are different sentences, and anyone quoting one as gospel is rounding away the uncertainty. the safe claim is the one i will make: as of late may 2026, anthropic is the most valuably-priced private ai company, and it got there fast. the reporting has it roughly tripling from a $380 billion mark in february. the part that matters more to me is the revenue. anthropic crossed a $47 billion run-rate earlier in may. that is the line that turns a valuation from a vibe into something with a floor under it. you can argue about whether $

You're about to install a VS Code extension. Maybe it's a formatter, a linter, a theme, an AI tool....

A free model that runs 4x faster on your own GPU — and two more shifts for builders Three things landed for builders at once: a free open model that generates text far faster, a more autonomous Codex, and Anthropic owning up to a model that was quietly holding back. Two of them you can act on right now. Here's the 2-minute video version if you want the quick pass first: 1. Google shipped DiffusionGemma — a free open model that runs 4x faster Google released DiffusionGemma , an open-weights model that uses text diffusion instead of standard autoregressive decoding. Instead of generating one token at a time, it generates whole blocks in parallel. It writes blocks of 256 tokens at once , for up to 4x faster generation on a dedicated GPU. It hits 700+ tokens per second on a single RTX 5090 , and fits in 18GB of VRAM quantized — inside consumer GPU limits. It's a 26B Mixture-of-Experts (only 3.8B parameters active), ships under Apache 2.0 , and runs natively in vLLM . The tradeoff Google states openly: output quality is lower than standard Gemma 4, so it's a speed play, not a quality play. Why it matters: this is a fast, free, local draft model you can run on your own hardware. Use it for low-latency drafts and agent loops, then route the hard calls to a stronger model. No inference bill for the cheap 80%. 2. OpenAI gave Codex web search and autonomous goals OpenAI shipped a major Codex update that pushes it further toward an autonomous agent. Code mode can now call web search directly , even from nested JavaScript tool calls — so it can look up current API docs mid-implementation. Goal mode is generally available across the Codex app, the IDE extension, and the CLI. Appshots (macOS) attach an app window to a Codex thread with a hotkey, and MCP tool schemas now preserve oneOf / allOf for richer connectors. Why it matters: Codex can research and chase a goal on its own across every surface. Still — hand it a clear, scoped goal in a branch. Full hand-offs go sideways witho

Every tool sprawl thread I read starts with license math, and license math is a decoy. Last quarter I added up what our seven-tool delivery stack actually cost us, and the subscriptions came to about 15% of the total. The other 85% never appears on an invoice, which is exactly why nobody budgets for it and nobody fixes it. Some background so you can judge whether my numbers transfer to your team. I spent years building automation in banking before running my own product team, so I am professionally allergic to process waste. Despite that, our stack had drifted into the usual shape: Jira for tickets, Confluence for docs, Lucidchart for architecture, TestRail for test cases, two spreadsheets doing unpaid overtime in the gaps, and an AI chatbot bolted on the side that had never seen any of it. The licenses for all of that, for six people, ran about $700 a month. Annoying. Not a crisis. And that is precisely why the "consolidate your tools" pitch dies in so many budget conversations. Saving a few hundred dollars a month does not justify a migration, and everyone in the room knows it. If licenses were the real cost, I would side with the skeptics. The audit: two weeks of logging every re-key So we measured the part nobody measures. For two weeks, everyone on the team logged every re-key: any moment a human moved or restated information that already existed in another tool. Copying acceptance criteria from Confluence into a Jira ticket. Updating TestRail because a story changed shape. Redrawing a Lucidchart flow that had drifted from the code. Reassembling a status update by hand from three tabs. Pasting project context into the chatbot, again, because it forgot everything since yesterday. The rules were strict so the number would survive scrutiny. Log transfer time only, not thinking time. Round down when unsure. If the same fact got re-keyed twice, log it twice, because it cost twice. Each entry went into a shared CSV with four columns, and this script turned it into th

AI coding tools are changing how software gets built. Claude Code, Cursor, GitHub Copilot, Windsurf and other tools can generate code incredibly fast. For small tasks, they are already useful: write a component, explain a function, scaffold an endpoint, create a test, refactor a file. But after using AI in real projects, one thing becomes obvious: The problem is no longer code generation. The problem is engineering control. Most AI coding workflows still look like this: text idea -> prompt -> code -> fix -> prompt again -> more code -> lost context -> start over It feels fast at the beginning. Then the project grows. Requirements change. Architecture decisions disappear inside chat history. The AI forgets previous context. You start acting as product manager, architect, reviewer, QA, DevOps, and prompt engineer at the same time. That is not software engineering. That is vibe coding. ## Vibe coding works until it doesn't Direct AI coding is great when the task is isolated. Ask for a React component. Ask for a SQL query. Ask for a utility function. Ask for a unit test. No problem. But real software is not a collection of isolated snippets. Real software has: - business rules - architectural constraints - existing patterns - security concerns - database impact - deployment requirements - edge cases - regression risk - long-term maintenance When AI jumps directly from prompt to code, it often skips the thinking that should happen before implementation. The result may compile. But does it fit the architecture? Does it respect the domain? Does it create hidden technical debt? Does it solve the right problem? That is the gap we are trying to close with N45.AI. ## What is N45.AI? N45.AI is a framework that turns AI coding tools into a structured engineering workflow. It works with the tools developers already use, including Claude Code, Cursor, GitHub Copilot, and Windsurf. The idea is simple: Instead of treating AI as one generic assistant, N45.AI organizes the work like a