标签：#Go

For the 14th time, a Google product has hit 1 billion users. Google CEO Sundar Pichai posted on X that a billion people are using Gemini every month, and that Gemini is Google's fastest-growing product ever. A billion users is a huge milestone, but Google isn't the first AI app to hit it. OpenAI's ChatGPT […]

Most HTML-to-Markdown tools handle one file at a time. You paste some HTML, get Markdown back, repeat. That works for a quick snippet but not when you have 200+ pages from a help center export sitting in a folder. I needed exactly that. I had a full site mirror (grabbed with wget --mirror ) and wanted clean Markdown I could feed into an LLM knowledge base. Nothing I found could handle it without uploading files to a server or converting one by one. So I built HTML to Markdown AI . How it works You drop a ZIP file (or individual HTML files) into the browser A Go-based conversion pipeline compiled to WebAssembly processes everything locally You get a ZIP back with clean GitHub-Flavored Markdown, folder structure preserved No server involved. Your files never leave your machine. The conversion pipeline The heavy lifting happens in Go/WASM. The pipeline: Strips navigation, footers, scripts, styles, and other boilerplate noise Extracts the main content from the page Converts to GFM with proper heading hierarchy, tables, code blocks, and links Handles batch processing so you can throw hundreds of files at it Why no built-in crawler? Intentional decision. Downloading HTML from someone else's site has legal implications depending on jurisdiction and terms of service. I don't want to be in that business. Downloading is also the easy part: wget -r -l 0 -np -k -E -p -e robots = off \ --reject-regex '\.(png|jpe?g|gif|svg|webp|woff2?|ttf|css|js|zip|pdf)$' \ -w 0.5 --random-wait \ https://docs.example.com/ That gives you a local folder with all the HTML. The hard and annoying part is turning that into clean, usable Markdown. That's what this tool solves. Stack Frontend: Astro + Tailwind Conversion engine: Go compiled to WebAssembly Processing: Entirely client-side, zero backend Try it https://www.html-to-markdown-ai.com Use cases I've tested it with: Help center exports (Zendesk, Confluence, custom wikis) Documentation sites mirrored with wget/httrack Scraped content for RAG pipe

Gemini is keeping pace with OpenAI’s ChatGPT, which hit 1 billion monthly active users back in June.

Google TV Freeplay, the company's free, ad-supported streaming service, now supports video on demand. Instead of tuning into Google TV Freeplay's selection of always-on channels, you can now choose from over 10,000 shows and movies to watch whenever you want. The update introduces titles like Lady Bird, Seventeen Again, and Hell's Kitchen, according to Google's […]

Scaleup Europe, a public-private fund with a $5.7 billion target, made its first investment by backing Finnish satellite company ICEYE.

Google is gearing up to reveal a bunch of new Pixel devices on August 12th. A series of leaks leading up to the event suggest that the Pixel 11 lineup will come in an array of colors, with signs pointing to a built-in light coming to Pro models. Teasers from Google also indicate that it […]

Last year's Made by Google event for the Pixel 10 resembled a late night talk show, complete with Jimmy Fallon as host, and it aired an hour after the phones were revealed. Spec dives and executive appearances were replaced with celebrity cameos and awkward skits, to the confusion of many diehard Pixel fans. But it […]

Claire Stapleton's new book, Don't Be Evil , recounts her journey from company cheerleader to outspoken dissident.

When I give talks about AI genies , I use this sort of example as a hypothetical. It’s happened . The story is from Australia. Someone named Andrew tasked OpenClaw to book gym classes for him. And…. Minutes later, his AI agent reported it had discovered a way to book Andrew into classes several weeks in advance, far beyond what was supposed to be possible. Andrew, who was sitting fourth on a waitlist for a class later that week, asked if it was possible to move him to the top of the list. The agent came back and told Andrew that it had kicked another gym-goer off the list as part of the testing of its capabilities...

The Quest Begins (The "Why") I still remember the first time I faced the “Next Greater Element” interview question. The array looked innocent enough, but every brute‑force attempt felt like I was hammering a nail with a sponge— O(n²) time, nested loops, and a sinking feeling that I was missing something elegant. I spent an hour sketching out the problem on a whiteboard, muttering, “There has to be a way to look ahead without looking back every single time.” That frustration is a rite of passage for many developers. We’re taught to think in terms of scanning left‑to‑right, but some array puzzles scream for a different perspective: we need to remember what we’ve seen in a way that lets us answer questions about the future elements instantly. Enter the monotonic stack—a deceptively simple data structure that turns those scary “look‑ahead” problems into straight‑line walks. The Revelation (The Insight) So what’s the secret sauce? A monotonic stack is just a stack that maintains its elements in strictly increasing or strictly decreasing order. Why does that help? Consider the Next Greater Element problem: for each index i , we want the first element to its right that’s larger than arr[i] . If we walk from left to right and keep a stack of indices whose next greater element we haven’t found yet, the stack will naturally be decreasing in value. Why decreasing? Imagine the stack holds indices [i₁, i₂, …, i_k] where arr[i₁] > arr[i₂] > … > arr[i_k] . When we encounter a new value arr[j] , any element on the stack that is smaller than arr[j] has just found its next greater element—namely arr[j] . We pop those indices, record the answer, and stop when we hit a value that’s not smaller (or the stack empties). Then we push j onto the stack. Because each index is pushed once and popped at most once , the total work is linear: O(n) . No nested loops, no repeated scans—just a single pass with a stack that does the heavy lifting. The same invariant works for other “first bigger/smal

AI jako partner, ne kalkulačka: člověk a AI při řešení Project Euler #185 srpna 2026 Co se stane, když člověk nepoužije umělou inteligenci pouze jako nástroj, který má dodat hotovou odpověď, ale jako partnera při řešení problému? Dnes jsme to vyzkoušeli na konkrétním problému z Project Euleru. Nechtěli jsme vytvořit nový algoritmus. Chtěli jsme zjistit, jak může vypadat skutečná spolupráce člověka a AI při hledání řešení. Experiment Vybrali jsme Project Euler #185 – Number Mind. Úloha obsahuje 22 šestnáctimístných sekvencí. U každé je uvedeno, kolik číslic je na správné pozici. Úkolem je najít unikátní šestnáctimístnou sekvenci, která splňuje všechna tato omezení. Na začátku jsme si stanovili jednoduché pravidlo: Nechceme pouze získat výsledek. Chceme společně hledat cestu k němu. První problém Naše první společná zkouška nedopadla podle očekávání. Ukázalo se, že jsme si pro experiment nezvolili ideální problém a postup. Místo toho, abychom se snažili chybu zakrýt, označili jsme první pokus jako neúspěšný a změnili postup. To se ukázalo jako důležitá součást experimentu. Chyba nebyla důvodem ukončit spolupráci. Byla informací pro další krok. Project Euler #185 U samotného problému jsme postupovali bez předem připraveného algoritmu. AI začala pracovat s kandidáty a jednotlivými řádky. Člověk průběžně sledoval strukturu problému a hledal jiný pohled. V určitém okamžiku přišel klíčový návrh: «„Nehledejme jen to, co je správně. Hledejme miny – čísla, která se nám nehodí.“» Tím se změnila orientace řešení. Místo hledání správných možností jsme začali systematicky vyřazovat možnosti, které nemohou být správné. Co přinesl člověk a co AI? Martin přinesl především: intuitivní pozorování, změnu perspektivy, rozhodování o směru dalšího řešení, pochybnosti a kontrolu jednotlivých kroků, myšlenku „min“. AI přinesla: rychlé zpracování velkého množství kombinací, strukturování hypotéz, systematické porovnávání, práci s omezeními, závěrečné ověření. Role se přitom během řešení nemě

Support on GitHub: github.com/phero20/concurrent-resource-scheduler (Give it a star if you find it useful!) View Docs: pkg.go.dev/github.com/phero20/concurrent-resource-scheduler What happens when thousands of concurrent requests compete for a small pool of reusable resources? You can put a mutex around a slice and hope for the best. Or you can design the scheduler around concurrency from the beginning. I chose the second option. I built Concurrent Resource Scheduler (CRS) , a domain-agnostic Go library for selecting, prioritizing, routing, and maintaining reusable resources under heavy concurrent load. The core idea is simple: MANY CONCURRENT REQUESTS │ ▼ ┌───────────────────┐ │ Resource Scheduler│ └─────────┬─────────┘ │ ┌──────────────┼──────────────┐ │ │ │ ▼ ▼ ▼ Priority Acquire State Heap Strategy Management │ │ │ └──────────────┼──────────────┘ │ ▼ BEST AVAILABLE RESOURCE But making that work correctly under concurrency is where things get interesting. CRS is designed for use cases such as: LLM/API gateways API key pools proxy rotation database replicas GPU workers backend pools worker resources connection pools rate-limited providers reusable compute resources The scheduler itself does not know what a resource means. It only knows: "I have resources. I need to safely maintain them, prioritize them, and return an appropriate one to a concurrent caller." Table of Contents The Problem The Naive Approach Why a Global Mutex Becomes a Problem The Core Idea Behind CRS Architecture at a Glance Sharded Priority Heaps Why Sharding Helps The O(1) Lookup Map Priority and Acquire Are Different Problems Acquire Strategies Round Robin Weighted Acquire Adaptive Acquire Affinity Routing Shared vs Exclusive Acquisition Resource Lifecycle Atomic State Transitions The Inactive Store Batch Operations Updates Without Destroying Heap Ordering Cooldowns Asynchronous Events Observability Prometheus Integration Concurrency Model Complexity Testing the Library Race Detector Validation

Fusion power startups are turning to Kyoto Fusioneering to supply components for future power pants. The Japan-based startup just received a grant to build a part of the fuel system.

The investigation shows that North Koreans are able to infiltrate government agencies, as well as private organizations and crypto exchanges.

Earlier this summer, Amazon customers began noticing that emails related to their online orders looked sparse: Order confirmation emails didn't name specific items anymore, and instead listed only item categories. "Your Beauty item is confirmed!" an email about my retainer cleaning tablets read. Shoppers have posted other iterations of the redacted emails as well: "Ordered: […]

Why passkey apps treat Windows differently than other operating systems.

Interesting empirical research: “ Black Box Warfare: Human Judgment and Military Decision-Making in the Age of AI .” Abstract: How is AI transforming decision-making in modern conflict? This study provides a unique empirical window into that question by deploying a high-fidelity replica of an AI decision-support system (DSS) used in military targeting. After reconstructing the interface and functionality of the real-world system, we tested its impact on combat decisions in two experiments involving 2,015 Israeli military personnel. Contrary to widespread fears of automation bias, we find strong evidence of algorithmic aversion, especially in scenarios involving high collateral damage. Yet we also show that integrating “explainable AI” features reduces algorithmic aversion and promotes more thoughtful evaluations of algorithmic recommendations. These findings challenge prevailing assumptions, revealing that trust in military AI is dynamic, varying with individual predispositions, perceived operational stakes, and the informational features of the interface. By grounding normative concerns in empirical evidence, our study offers critical insight into the integration of AI in warfare and underscores the enduring importance of human agency in high-stakes military decision-making...

Hi everyone I’ve been hacking on a terminal coding agent called phi https://github.com/pulseaiclub/phi for the past few months, and I’d love to share what it is and why it exists. If you’ve used tools like Pi, Claude Code, Aider, or Goose, phi tries to hit the same workflow—but deliberately strips away the runtime baggage. It’s a single Go binary, ~12 MB, with no Node, Electron, or Python in sight. Here’s what makes it tick: No model lock-in This is the whole reason phi exists. Model releases are moving faster than any agent maintainer can keep up with. Instead of chasing every new provider, phi tr eats the model as a pluggable config: anything OpenAI-compatible or Anthropic-native works out of the box. You can swap models in seconds without waiting for a n author to add support. Config is a single YAML file, and there’s also a tiny HTML editor so you can point-and-click your way through ~/.phi/config.yaml. As light as it gets Go is the secret sauce here. A stripped release build (CGO_ENABLED=0) comes in at ~12 MB, cold idle RSS around ~21 MB, and a first frame in roughly 40 ms. Ther e are only 6 direct module dependencies. If you want a coding agent you can go build in half a second and read in an afternoon, this is it. A permission gate that actually matters One thing that always makes me nervous with coding agents is the "model deletes your files" moment. Phi defaults to read-only: the agent can scan your codebase , but writes and shell commands require explicit approval. The approval dialog gives you three choices—allow, deny with feedback, or allow everything for the r est of the session. Rules are granular: • bash.allow → go test ./... is fine. • bash.deny → rm -rf * is blocked. • fetch.allowed_hosts → restrict which domains the agent can reach. Sub-agents that don’t bloat your context One big agent is fine for small tasks, but long-running work pollutes the context window with noise. Phi ships a full set of sub-agent tools (agent_spawn, agen t_task, agent_wai

The Quest Begins (The "Why") I still remember the first time I saw LeetCode 323 “Number of Connected Components in an Undirected Graph”. I stared at the adjacency list, thought “I’ll just run a DFS from every node”, and coded it up in ten minutes. The solution passed the easy tests, but when the hidden test cases hit a graph with 10⁵ nodes and 10⁵ edges, my DFS started to choke—stack overflows, repeated visits, and a sinking feeling that I was brute‑forcing a problem that deserved a smarter tool. That night, after a few too many coffees, I stumbled upon a tiny comment in a discussion thread: “Union‑Find can do this in almost O(1) per operation”. My curiosity sparked like a power‑up in a retro arcade game. I had to know why this seemingly simple data structure could turn a nightmare into a breeze. The Revelation (The Insight) At its heart, Union‑Find (aka Disjoint Set Union, DSU) maintains a collection of elements partitioned into disjoint subsets. It supports two operations: Find(x) – returns the representative (root) of the set containing x . Union(x, y) – merges the sets containing x and y . The magic lies in two simple heuristics: Path Compression – when we walk up the tree to find a root, we make every node on that path point directly to the root. Future finds become flat, almost constant‑time. Union by Rank/Size – we always attach the smaller tree under the root of the larger one, keeping the overall tree shallow. Why does this give us near‑O(1) amortized time? Think of each Find as paying a small “tax” to flatten the path. The tax is paid only a few times per node before it becomes a direct child of the root. Over a sequence of m operations, the total work is bounded by O(m α(n)) , where α is the inverse Ackermann function—so slow‑growing it’s practically a constant for any realistic n . In plain English: every time we climb up, we leave a shortcut behind. The next climber benefits from that shortcut, and the structure keeps getting better. It’s like building

If you love the simplicity of Docker Swarm (native Compose files, lightweight single binary) but miss the advanced capabilities of Kubernetes (targeted label placement, SRE-grade observability, built-in DNS service discovery, and zero-trust ingress), meet Gubernator (gbnt) . We are excited to release Gubernator v2.13.0 , introducing three massive feature suites natively integrated into a single binary and a modern Material Design 3 Flutter Web Dashboard: Google SRE Multi-Burn-Rate SLO Engine & Interactive Suite CoreDNS 4-Tab Management Suite & Interactive Dig Playground Caddy Ingress & Zero-Trust Reverse Proxy Suite Fun Fact: The entirety of Gubernator's codebase, multi-node deployment pipelines, and SRE features were designed, built, and pair-programmed using **Google Antigravity (AGY) , Google DeepMind's agentic AI coding assistant! Let's dive into what's new and how you can level up your self-hosted or production container clusters! 1. Google SRE Multi-Burn-Rate SLO Engine & Web Suite Defining Service Level Objectives (SLOs) and tracking Error Budgets is the gold standard of Site Reliability Engineering. Until now, implementing SLOs meant running heavy Kubernetes CRDs (via tools like Sloth or Pyrra) or using costly SaaS platforms. Gubernator v2.13.0 brings Google SRE Workbook (Chapter 5) compliant multi-burn-rate alerting straight to simple docker-compose.yml services: version : " 3.8" services : payment-api : image : hashicorp/http-echo:latest labels : gbnt.slo.enable : " true" gbnt.slo.target : " 99.9" gbnt.slo.window : " 30d" gbnt.slo.template : " caddy-http" gbnt.slo.journey : " Checkout Flow" What makes Gubernator's SLO Suite unique? Google Multi-Burn-Rate Alerting : Automatically generates standard 4-window Prometheus recording and alert rules ( Critical Page 1h/6h & Warning Ticket 3d/14d ). Dynamic "No-Code" Management : Click "+ Configure / Add SLO" in the Web UI or call POST /v1/slo/edit to create, edit, or disable SLOs on the fly without editing Compose