标签：#t

Stack Overflow's question volume has been falling since ChatGPT went public in November 2022 ( OpenAI ). The site that trained a generation of developers, and most of the AI tools those developers now use, is slowly emptying out. In October 2023, Stack Overflow laid off 28% of its staff ( Stack Overflow Blog ). CEO Prashanth Chandrasekar framed it as a restructuring toward profitability. Everyone in the industry understood the real cause. Traffic was down. The thing causing it was sitting in every developer's browser tab. This is not another "AI killed Stack Overflow" piece. That take is everywhere and it misses the actual problem. The interesting part is the feedback loop, and it points somewhere uncomfortable for the AI industry itself. The conventional story, and what it misses The popular version goes like this. Developers used to paste error messages into Google and land on a Stack Overflow thread. Now they paste the same error into ChatGPT, Claude, or Copilot and get a direct answer. Why click through to a forum, risk a condescending comment, and wait for a human when a model answers in two seconds? That part is true. It explains the traffic drop. It does not explain why the people building the AI should be worried. The seed corn problem Here is the part most coverage skips. Every large language model trained on internet text consumed a huge amount of Stack Overflow. The site's archive of voted, edited, human-reviewed answers is one of the highest-quality programming datasets in existence. It is the reason an AI can answer your Python error at all. Now run the loop forward. AI tools answer questions directly. Developers stop posting on Stack Overflow. The archive stops growing. The next round of models trains on a corpus that is increasingly old, increasingly stale, and missing everything that happened after 2022. When you train an AI on data generated by another AI, quality degrades. Researchers proved this formally. Shumailov and colleagues showed that model

I watched an AI agent send 200 emails to the wrong recipients because I forgot one validation check. The emails were well written. The offers were real. The recipients were just... not our leads. That was early. I learned fast. Every agent I build now has three layers of guardrails before it touches a database or an API. Here's exactly what those layers look like and why they're non-negotiable for production. Input Validation: Your Prompt Is Not a Schema The first mistake people make is trusting the LLM to produce valid output. It won't. Not reliably. I've seen GPT-4 return a JSON key called "emial" instead of "email" in a critical pipeline. One typo, and the whole record is garbage. The fix is a strict validation layer that runs before any data reaches your system. In my AI resume tailor, I use a JSON schema with conditional presence flags. Every field that must be real has a has_* boolean guard. If the LLM tries to fabricate a phone number, the schema rejects it. const resumeSchema = z . object ({ contact : z . object ({ email : z . string (). email (), phone : z . string (). optional (), has_phone : z . boolean () }). refine ( data => { // If phone is present, the guard must be true return data . phone ? data . has_phone : ! data . has_phone }, " Phone number present but has_phone flag is false " ) }) This pattern catches hallucinations before they corrupt your database. The schema is the contract. The LLM is just a suggestion engine. Permission Scoping: Give Agents the Minimum They Need An agent should never have write access to tables it doesn't need. That sounds obvious, but I've seen production systems where a job description rewriting agent had full CRUD access to the user table. When I built the LLM scoring pipeline for a job board platform, I created separate database roles. The scoring agent only had SELECT on the job listings table and INSERT on a scoring results table. It never touched users, applications, or configuration. Even if the prompt was hijack

Originally published on hexisteme notes . I write these engineering notes with one main model, and until recently I also reviewed them with that same model. Same family writes, same family checks its own work. That sounded fine right up until I had ten queued posts sitting in a publish backlog and a nagging thought: if the writer and the reviewer come from the same training distribution, what exactly is the review checking for? So I ran an experiment. I took the queue and had a different vendor's model audit it before anything went out — not to replace my own review, but to see what a genuinely different set of weights would flag that mine hadn't. The setup: copies only, and a self-verifying prompt The mechanics were deliberately boring. I copied the ten queued articles into a scratchpad directory and exposed only that copy to the auditor via --add-dir — the auditor never got write access to the originals, so nothing it did could touch the source of truth by accident. The audit itself ran as agy --model gemini-3.1-pro-high , pointed at the copy directory, with one instruction: find technical factual errors, broken sentences, cross-article inconsistencies, unsupported claims, and tone violations, and verify each one yourself on the web before reporting it. I wanted a model that would check its own homework, not just pattern-match on "this looks wrong." It came back with seven findings. Rule one: don't trust the auditor either Seven findings from a different vendor is not the same thing as seven confirmed bugs. I re-verified every single one independently — grepping the original text, checking official documentation, and where possible checking against a real machine — before touching anything. Of the seven, six held up and got fixed. One didn't: the auditor flagged a sentence as an error, and when I went back to the primary source, it turned out to be the auditor misreading a perfectly correct sentence, not a defect in the writing. Without the re-verification step, I

The end of the FIFA Men’s World Cup is nigh. Here’s how to watch the final games and the first ever World Cup halftime show.

Organizers want the show to be the most-viewed halftime show in history. Having Shakira, Justin Bieber, Coldplay, and BTS will surely help—but do soccer fans want it?

Keep your desktop clean by auto-organizing screenshots. Discussion | Link

Most C# developers know how to use yield return. Few understand what actually happens after compilation. If you've ever written something like this: public IEnumerable < int > GetNumbers () { yield return 1 ; yield return 2 ; yield return 3 ; } it looks almost magical. No collection. No list allocation. No iterator implementation. Yet somehow the method returns an IEnumerable. So what is really happening? The answer is one of the most elegant compiler transformations in the entire .NET ecosystem. Let's open the hood. The Illusion Most developers imagine the previous code executes like this: Call method ↓ Return 1 ↓ Pause ↓ Return 2 ↓ Pause ↓ Return 3 That isn't what happens. C# methods cannot actually pause execution. Instead, the compiler completely rewrites your method into something entirely different. The Compiler Creates a State Machine Your tiny method becomes a hidden class similar to this: private sealed class GetNumbersIterator : IEnumerable < int >, IEnumerator < int > { private int _state ; private int _current ; public bool MoveNext () { switch ( _state ) { case 0 : _current = 1 ; _state = 1 ; return true ; case 1 : _current = 2 ; _state = 2 ; return true ; case 2 : _current = 3 ; _state = - 1 ; return true ; default : return false ; } } public int Current => _current ; } Your original method no longer exists. Instead, it simply returns: return new GetNumbersIterator (); Every yield return becomes another state inside MoveNext(). Why Local Variables Don't Disappear Consider this code: IEnumerable < int > Squares () { int x = 1 ; while ( x <= 3 ) { yield return x * x ; x ++; } } After the first yield, the method "pauses." But where is x stored? Not on the stack. The original stack frame has already disappeared. Instead, the compiler promotes local variables into fields: private int _x ; The iterator object now owns every variable that must survive between iterations. This is why iterator methods can remember where they left off. Heap Allocation Happens Ma

Company brain for AI Agents Discussion | Link

When I first wrote about NutriAgent in November 2025, it was a full application. It had a Python backend, a web interface, a Telegram bot, user accounts, Google OAuth, Supabase, and a Google Sheets integration. Recently, I reproduced its core workflow as a skill for my personal AI agent. It took around 15 minutes and two prompts. I didn't build another backend, deploy a service, or implement OAuth again. I explained how I wanted the workflow to behave, tested it, and watched a new row appear in my nutrition spreadsheet. The original application wasn't a mistake. It was how I could deliver that experience with the tools available at the time. Eight months later, the starting point had changed. Eight Months Ago, This Was an App I built NutriAgent because I wanted to track calories and protein without trapping my data inside a nutrition app. I wanted the raw records in a spreadsheet I controlled, where I could create my own reports and eventually connect nutrition with my training data. The first version was a personal n8n workflow. It worked for me, but when a friend wanted to try it, I realized that everything was tied to my accounts. To make it reusable, I rebuilt it in Python and added the parts a real multi-user product needed: authentication, storage, a web interface, Telegram, Google OAuth, conversation history, and account linking. I've already told that story in I Ditched MyFitnessPal and Built an AI Agent to Track My Food , and later wrote about what broke after I used it every day for a month . This article starts after that version. My Gaming PC Became an Agent Box I had a modest gaming PC at home with 16 GB of RAM and a 1 TB drive. Using it through Windows, WSL, and remote desktop from my Mac felt awkward, so I installed Linux and turned it into a remote box for running agents. I'll write about that setup separately. I moved Hermes there from a VPS. Hermes is the personal agent I now run on that machine. It can load reusable skills and use tools connected

A terminal, a real browser, and Claude Code under one roof Discussion | Link

Category theory often sounds like impenetrable academic jargon, but at its core, it is simply the mathematics of composition. In functional programming, we use its concepts to build modular, predictable, and bug-resistant code. JavaScript isn't a pure functional language like Haskell, but it has first-class functions and treats functions as values. This makes it entirely possible to implement category theory primitives. Here is how the theoretical concepts map to practical JavaScript. The Vocabulary of Category Theory Before jumping into code, let's define the fundamental pieces: Categories: A collection of objects and arrows (morphisms) between them. Objects: In programming, these are our types or data (e.g., String, Number, Boolean, Array). Morphisms: These are our pure functions that transform one type into another (e.g., a function that takes a String and returns a Number). The goal of functional programming primitives is to create safe wrappers (containers) around our data so we can compose these functions predictably, without side effects or unhandled errors. 1. Functors: The Mappables A Functor is any type that implements a map method. It is a container holding a value, and map allows you to apply a function to that value without pulling it out of the container. When you map over a Functor, it returns a new Functor of the same type, preserving the container's structure. The native JS Functor: You already use Functors every day. JavaScript Arrays are Functors. const numbers = [ 1 , 2 , 3 ]; // We apply a function to the values inside, and get a new Array back const doubled = numbers . map ( x => x * 2 ); // [2, 4, 6] Building a custom Functor: Let's build a simple Box container to see how this works for single values. const Box = x => ({ map : f => Box ( f ( x )), fold : f => f ( x ), // An escape hatch to get the value out inspect : () => `Box( ${ x } )` }); // Usage: const result = Box ( ' Functional Programming ' ) . map ( str => str . trim ()) . map ( str

Originally published at blog.whynext.app . It started as an ordinary cleanup problem. Users upload media files (recordings and images) through presigned URLs. The server issues an upload URL, the client uploads straight to S3, then calls a commit API to say "register this key as a real asset." The problem is what happens when someone gets a presign but never commits. The app crashes, the network drops, the user leaves the screen, and the bucket is left with an object that isn't registered anywhere. I wanted a lifecycle rule to clean these up, but there was no way to write one. Committed and uncommitted objects were mixed under the same prefix, so any rule that says "delete old things" would delete real assets too. A daily upload quota kept the pile from growing fast, but the fact remained: there was no path to reclaim the space. The design: what isn't committed lives in tmp The backbone of the fix is key namespace separation. presign issues a temporary key under the tmp/ prefix. When commit passes validation (existence check via HEAD, Content-Type, size limit), it promotes the object to its final key with CopyObject and deletes the tmp original. Objects whose commit never arrives stay in tmp/ , and a lifecycle rule expires them after 7 days. Now the lifecycle rule only has to look at tmp/ . Real assets are outside its blast radius from the start. The clients didn't need to change. I read all three upload flows to confirm this: every one of them uses the key returned in the commit response for its follow-up calls, so the server can change the key shape without them noticing. Commits for old-format keys already in flight at deploy time still go through the existing path. One trap here. This bucket has versioning enabled. On a versioned bucket, expiration doesn't delete an object. It only adds a delete marker, and the original bytes stay behind as a noncurrent version. Without a paired noncurrent_version_expiration (1 day), the cleanup runs and not a single byte is rec

AI coding agents have dramatically shortened the distance between an idea and working software. They can inspect a project, create files, run commands, write tests, and help diagnose failures. What they have not eliminated is judgment. A polished screen is not proof that data survives a reload. A passing unit test is not proof that keyboard users can complete the core task. A successful deployment is not proof that the intended commit reached production. This is why I use proof gates : observable conditions that must be satisfied before a product claim becomes stronger. A proof gate is not a meeting, a long document, or an excuse to slow down. It is a compact question: What evidence would let another person verify that this claim is true? Here are five gates that separate a persuasive AI demo from a small MVP you can responsibly ship. Gate 1: Prove One Valuable User Loop AI makes feature generation cheap, which makes uncontrolled scope especially dangerous. Before requesting code, define one primary user in one specific situation. Then describe: Their observable before-state The smallest useful action they can take The immediate result The reason they might return This becomes the product’s core loop. For example, “build a productivity platform” is too broad. A more testable loop might be: A freelancer remembers a useful client outcome. They record the outcome and supporting evidence. The record appears in a searchable library. They can retrieve it later for a proposal or review. The gate is not passed because a form exists. It is passed when a new user can complete the entire loop and explain what changed without coaching. Write a Not Today list alongside the required capabilities. Authentication, dashboards, collaboration, billing, and AI-generated summaries may all be reasonable later. They should not compete with proof of the first useful loop. The goal is not the fewest possible features. It is the smallest complete behavior that tests whether the product creat

Originally published at boldpilot.club Most startup launches fail not because the product is wrong, but because the planning collapses under its own ambiguity. A solid launch plan template for startups gives you a structure that converts vague intentions — "we'll do some PR, post on social, maybe a Product Hunt thing" — into sequenced tasks with owners and deadlines. If you're building something and your launch plan currently lives as a notes app dump or a half-filled Notion page with seventeen unticked checkboxes, that's the problem this article addresses directly. A usable launch template has five core components: a pre-launch timeline, a channel priority stack, a messaging framework, a metrics baseline, and a post-launch review cadence. Everything else is optional decoration. Why Most Launch Plan Templates Get Abandoned The templates people download from Google — the ones with seventeen tabs and color-coded Gantt charts — get opened once and then quietly archived. I've seen this happen with teams that had more than enough time and resources. The issue is that those templates are designed to look thorough, not to be used under the actual pressure of a launch week. According to CB Insights, 35% of startup failures cite "no market need" as the primary cause, but the second and third most common reasons — running out of cash and not having the right team — are both downstream of poor planning and sequencing. A launch plan that nobody opens doesn't prevent any of those outcomes. The templates that do get used share one quality: they're shallow enough to skim at 11pm the night before a deadline. One page if possible. Two at most. If your template requires a tutorial to operate, it's already failed. I'd push back on the common advice to "start with goals." Goals are fine, but most founders already know what they want — signups, revenue, waitlist growth. What they're missing is the sequencing. Start with the constraint instead: what's your actual launch date, and what ab

Every OP Stack chain we've covered — Base, Unichain, Zora — has a sequencer: one privileged party that orders transactions, and the thing you're implicitly trusting for liveness and fair ordering. Taiko doesn't have one. It's a based rollup : Ethereum's own validators propose Taiko's blocks as part of normal L1 block production. That single architectural choice cascades into everything a developer cares about — liveness, finality, MEV, and reliability. And because Taiko is also a Type-1 zkEVM , your Ethereum tooling works with zero changes. Here's the map for chain ID 167000 . The essentials Taiko mainnet ( Alethia ) is chain ID 167000 , an EVM Layer 2 with: ETH as the gas token (18 decimals) — no separate gas token to source. ~12-second blocks , aligned with Ethereum's slot times — because block proposing rides on L1, the cadence follows L1. Type-1 zkEVM equivalence — the most Ethereum-equivalent zkEVM design. Contracts deploy bit-identically; opcode behavior is exact. Connecting is completely standard EVM: import { createPublicClient , http } from " viem " ; import { taiko } from " viem/chains " ; // chain ID 167000 const client = createPublicClient ({ chain : taiko , transport : http ( " https://rpc.swiftnodes.io/rpc/taiko?key=YOUR_API_KEY " ), }); await client . getBlockNumber (); // just works What "based" changes: no sequencer to trust — or to fail On a typical rollup, a sequencer receives your transactions, orders them, and produces L2 blocks ( what a sequencer does ). It's efficient, but it's also a single point of trust and a single point of failure — sequencer outages have taken major L2s offline for hours. A "based" rollup removes it entirely: Block proposing happens on Ethereum L1. Taiko blocks are proposed via L1 transactions, so Ethereum's proposers include them as part of normal block production. There is no separate Taiko sequencer. Liveness = Ethereum's liveness. As long as Ethereum is producing blocks, Taiko is producing blocks. There is no "the se

A video-to-prompt tool looks simple from the outside: upload a clip, wait a moment, and copy the result. The hard part is not generating text. It is preserving enough of the source video's structure that the prompt remains useful when another model interprets it. I learned this while working on a small video analysis workflow. Early versions produced fluent paragraphs, but they often dropped a camera move, merged two events, or placed dialogue in the wrong shot. The output sounded good and still failed as a production prompt. The fix was to stop treating the result as one block of prose. Start with an intermediate representation I now treat the prompt as the last stage of a compiler. The video is first converted into a structured record, and only then rendered for a specific video model. A minimal record might look like this: { "duration_seconds" : 12.4 , "shots" : [ { "start" : 0.0 , "end" : 3.8 , "subject" : "a cyclist waiting at a red light" , "action" : "looks over the left shoulder" , "camera" : { "shot_size" : "medium" , "movement" : "slow push-in" , "angle" : "eye level" }, "dialogue" : null } ] } This structure is deliberately boring. That is useful. A typed record makes missing data visible and gives you something concrete to validate before you ask a language model to write polished prose. Normalize the input first Video files arrive with different frame rates, codecs, orientations, and audio layouts. Links from social platforms add another layer of inconsistency. If every downstream stage has to understand every input format, failures become difficult to reproduce. The ingestion stage should produce a canonical package: a timestamped frame stream at a known sampling rate a normalized audio track basic metadata such as duration, aspect ratio, and frame rate a stable internal time base Keep the original timestamps. Rounding everything to whole seconds is tempting, but it causes trouble in short clips where several actions happen in quick succession. Detect

AWS has recently introduced CloudFormation express mode, a deployment option that can reduce infrastructure deployment times by marking stack operations complete once resource configuration is applied, rather than waiting for full resource stabilization. By Renato Losio

Everyone's talking about AI agents right now, but most tutorials stop at "call an LLM in a loop." If...

If you've ever localized an iOS or macOS app, you know the drill: you've got a .plist file full of UI strings sitting in Base.lproj , and now you need the same file in fr.lproj , th.lproj , de.lproj ... and so on for every language you support. You can pay for a translation service, hand it to a freelancer, or sit there manually retyping strings into fifteen copies of the same file. I didn't want to do any of those things, so I spent a day building pListTranslatorApp — a small SwiftUI macOS app that opens an Xcode plist, finds the string values you want translated, batch-translates them using Apple's built-in Translation framework, and writes out a ready-to-use localized plist. It's free, it runs entirely on-device, and it's now sitting on GitHub for anyone who wants it. The constraint that shaped the whole thing I'm developing on a MacBook Air with a 256GB SSD and 8GB of RAM — not a lot of headroom. That ruled out a couple of obvious approaches: Cloud translation APIs (DeepL, Google Cloud Translate, etc.) work well, but even a generous free tier means an external dependency, an API key to manage, and a network round-trip for something that's ultimately just short UI strings. Downloading a pile of on-device language models "just in case" eats real disk space fast — Apple's Translation framework models are shared system-wide across apps, which helps, but they still add up if you're not paying attention. So the app leans entirely on Apple's Translation framework , downloading exactly one language pack at a time, on demand, and lets you delete it once you're done with that language. What it actually does Open a plist. Pick any Xcode .plist via a standard file importer. Choose which keys to translate. Originally I hardcoded it to look for a key called itemTitle , but that's obviously too narrow for anyone else's project — so it's now a comma-separated text field. Type itemTitle, title, label and it'll pick up all of them, anywhere in the plist's nested dictionaries and

This is the workflow I follow before I use AI agents to implement any feature or bug fix. 🧭 Requirements/Specification ↓ Design/Architecture ↓ AI Code Generation ↓ Human Review ↓ Build & Static Analysis ↓ Testing & Validation ↓ Defect Resolution ↓ Security & Compliance Review ↓ Release ↓ Production Monitoring vs Claude Code ↓ Implements feature ↓ Codex QA Agent ↓ Runs application ↓ Tests happy path ↓ Tests edge cases ↓ Tests error handling ↓ Produces QA report This will resolve the self-review bias, confirmation bias, or AI-to-AI bias. 1️⃣ Understand Before Writing Code Before touching any code, I try to understand what I'm building and why . I usually start by reading: specs/<module>/<TICKET>-<slug>.md plan/<module>/<TICKET>-<slug>.md status.md Then I review the project conventions: specs/CONVENTIONS.md specs/conventions/core-porting.md Finally, I read the existing implementation (entities, services, mappers, etc.) so my changes follow the existing architecture instead of introducing a new style. 💡 Pro-Tip Good code fits into the codebase. Great code looks like it was always there. 2️⃣ Plan the Change Once I understand the requirements, I identify which architectural layers are affected. I always respect the dependency order: Schema / Entities / DAOs ↓ Mappers / DTOs ↓ Service Layer ↓ Application Layer ↓ Controllers I don't jump ahead of dependencies. If a change is complicated or ambiguous, I document the approach before writing code. --- ## 3️⃣ Write the Code While implementing, I follow the repository's rules. Some examples: | Rule | Detail |---|---|---| | DTOs | Generated from `schema.yml` — never handwritten | | Status values | Sourced only from the Core Porting specification | | Traceability | Every ported behavior includes a source citation | Citation formats I use: - `← Source <path>` - `← PS §...` - `← BR-###` Beyond repository rules, I also try to: - ✅ Match existing naming conventions - ✅ Keep comments minimal and meaningful - ✅ Make small, focused chang