标签：#AI

Natural gas prices could triple in some parts of the U.S., which could saddle hyperscalers with massive bills to power their AI data centers.

For years, I thought being a good programmer meant knowing your stack really well. I was a Laravel developer, A React developer, A Node.js developer and A Go developer. And there was some truth to that. I spent years working with Laravel, for example, and naturally became faster at solving problems with Laravel. I know the ecosystem, the common mistakes, the packages, the conventions, and probably a few things that weren't even written in the documentation. My stack became part of my identity as a developer. But I think AI is slowly changing that. Not because frameworks and programming languages don't matter anymore. They obviously do. It's because AI has made moving between them much easier. Today, I can open a codebase written in a language or framework I haven't touched in years, or maybe have never used seriously, and get productive much faster than I could before. I can ask AI to explain the project structure. I can ask it to explain a piece of code. I can ask it to translate something I understand in PHP into Go. I can ask it to help me write tests. I can use it while debugging. I can even ask it why a particular approach might be a bad idea. That doesn't suddenly make me an expert in that technology. But it means I don't need to spend weeks just getting comfortable enough to start solving the actual problem. And I think that's a pretty big change. Your Stack Is Becoming Less Important There was a time when knowing a technology itself was a significant advantage. If you knew Laravel, you had to learn Laravel. If you wanted to learn React, you had to spend time understanding React. If you wanted to work with Kubernetes, good luck. You read documentation, watched tutorials, built things, broke things, fixed them, and slowly built up experience. That's still how you become good. But AI has changed the entry point. The first few hours with a new technology are no longer as painful as they used to be. You can have an AI sitting beside you explaining things as you g

security, #api, #ai, #cybersecurity A green CI/CD build means almost nothing to a regulator. Your AI hiring tool can pass every unit test, lint rule, and license scan, and still ship training labels from a sanctioned data broker. Legal only has to ask one question to turn that green pipeline red: who screened the vendors? High-risk AI systems need more than accurate models. A sanctioned supplier can poison your training data, cloud bill, or payment rail. The failure is usually not negligence; it is that compliance checks live in spreadsheets while the code lives in Git. A CI-ready sanctions helper in 40 lines I wanted the check inside the same pipeline that runs pytest. This helper screens a list of names against all five major sanctions lists and prints a markdown report that the CI runner can fail on. import os import sys import requests API_KEY = os . getenv ( " RAPIDAPI_KEY " ) if not API_KEY : sys . exit ( " RAPIDAPI_KEY is not set " ) URL = " https://sanctions-screener.p.rapidapi.com/screen " HEADERS = { " X-RapidAPI-Key " : API_KEY , " X-RapidAPI-Host " : " sanctions-screener.p.rapidapi.com " , } def screen_name ( name : str ) -> dict : try : r = requests . get ( URL , headers = HEADERS , params = { " name " : name }, timeout = 10 , ) r . raise_for_status () return r . json () except requests . exceptions . Timeout : return { " error " : f " timeout for { name } " } except requests . exceptions . RequestException as e : return { " error " : f " request failed: { e } " } def print_report ( name : str , result : dict ) -> None : print ( f " ## { name } " ) if " error " in result : print ( f " **ERROR:** { result [ ' error ' ] } " ) return verdict = result . get ( " verdict " , " UNKNOWN " ) print ( f " **Verdict:** { verdict } " ) matches = result . get ( " matches " , []) if not matches : print ( " - No matches " ) return for hit in matches : field = hit . get ( " matched_field " , " unknown " ) mtype = hit . get ( " match_type " , " unknown " ) tokens = hit .

Learn how containers communicate with each other and how to keep data alive even after containers are removed. Modern applications rarely run as a single container. A typical application might include a web application, a database, a cache layer, and background workers. For these services to work together, containers need a reliable way to communicate and share data. In this article, we'll learn: How Docker networking works How containers discover each other Docker network drivers Persistent storage with Docker volumes Essential networking and volume commands A real-world multi-container example By the end, we'll understand two of the most important concepts in Docker: networking and data persistence . Why Docker Networking Matters Every container runs inside its own isolated network namespace. This isolation improves security and prevents conflicts, but it also creates an important challenge: If containers are isolated, how does a web application connect to a database? Imagine a web application running inside one container and MongoDB running inside another. Without networking, they cannot communicate. Docker solves this problem using Docker Networks . A Docker network allows containers to communicate with each other while remaining isolated from unrelated containers. Web App Container | v Docker Network | v Database Container Without a shared network, containers cannot easily find or communicate with each other. Docker Network Drivers Docker supports several network drivers, but most developers primarily use three. Bridge Network A bridge network creates a private virtual network on the Docker host. Containers connected to the same bridge network can communicate with each other securely. Create a custom bridge network: docker network create my-app-network Benefits of bridge networks: Container-to-container communication Isolation from other applications Built-in DNS resolution Easy management For most Docker projects, a user-defined bridge network is the recommend

Category: My AI Experiments There's a mental model most people use when working with AI: describe your problem, get a solution. It works well enough, until it doesn't. And when it fails, the failure is invisible — because AI doesn't say "I don't have enough to go on." It gives you a confident, well-reasoned, completely wrong answer. I learned this the hard way during a website security investigation. The AI and I ran a thorough analysis, reached a clear conclusion, and were wrong. Not because the AI was weak — because I gave it the wrong kind of input. When I changed the input, the same AI found the answer in seconds. That gap — between the input that produces a wrong answer and the input that produces a right one — is what I want to talk about. The Investigation That Almost Failed My website was secretly redirecting visitors to a virus site. The attack was sophisticated: it only targeted specific browsers, fired at most once per device per day using a cookie-based cooldown, and left no trace in any file. I asked AI to help investigate. I described the symptoms. We searched through files together — .htaccess , theme functions, plugin code. Everything looked clean. The AI identified the most suspicious external element in scope: a Chinese analytics script called 51.la. I removed it. The redirect stopped. I called it solved. Three weeks later, the identical attack appeared on another site I manage. No 51.la anywhere. This time, instead of describing the symptoms, I gave the AI something different: the actual rendered HTML of an affected page, fetched using the exact browser User-Agent and IP type that triggered the attack. The AI found an 83KB malicious JavaScript payload injected into every page. Inside it: a WeChat browser detector, a link-click hijacker, a cookie-based daily cooldown. The payload was stored in the WordPress database — in plugin configuration data — where no file-level search could ever find it. Same AI. Same type of problem. Completely different ou

On 1 August I opened an issue that was three sentences long. A hundred and one minutes later the feature was merged, and the document that got it there ran to 457 lines . I didn't write those 457 lines. In fact, I didn't have to write any more documentation, and not because I simply allowed Claude to run amok. Here is the issue in full — control-api#265 , 225 characters: control-api#265 — Manifest-backed dashboard feeds For each dashboard, auto create a manifest keyed by dashboard_id. For each sensor the dashboard uses, tag it to be included in the manifest. When a dashboard definition is updated, add / remove tags from sensors accordingly. From that genesis moment, this is the lifecycle of the issue all the way through to landing: Time (UTC) Event 14:25 Issue #265 opened — 225 characters 14:54 FEAT-0007 spec committed — 457 lines 15:35 Spec merged (PR #266) 15:51 Implementation committed 16:06 Implementation merged (PR #267, 15 files), issue closed The interesting part isn't the speed. It's the step at 14:54 that landed a previously non-existent spec document, and what happened in the twenty-nine minutes before it. The issue was never a specification I often write issues like this one...the way most people write shopping lists. Actuator address is not ensured? Baseline the trace correctly. With the pre-rolls, the frame-rate looks out. They're abbreviated to the point of being cryptic to everyone else. I write them this way deliberately: I'm usually mid-something else when I notice a problem, or have an idea for a better route to the solution. The cost of a full write-up right at that moment would be a fractured sense of flow. As most engineers will tell you, the transitions into and out of flow are the most disruptive parts of their working day. This terse form of issue-writing can be all you need, and it's worth being precise about why it works and the trade-offs it includes. It is not because "the issues are good enough". They aren't. When you pick one of these u

Times are changing, my role is changing, my focus is changing, my impact is changing. But in essence – I'm still doing the same. I still build products that drive impact. Only my day-to-day looks completely different. The shift is happening, sooner or later, if you want it or not. Whether or not you can cope, is all up to you. In the past, I was neck-deep in code. That was what the majority of my time consumed. I liked it, building things, building products. These days, that's all done by an endless amount of AI agents. I barely touched any code in the past half year – if not even longer. My focus moved from building products to building my own process The work that used to go into a feature now goes into the process that produces the feature. Instead of losing the first hour of my day to Slack and email, I built a small stack of scheduled agents that hand me a briefing before I even open my laptop ( already wrote about that one ). Instead of reading every pull request line by line, I set up a review loop where agents do the first pass and I stay on the hook for whatever they flag. None of it started as a plan. Each piece started as one specific annoyance I got tired of and fixed. That's the actual mechanism: improve one small thing, it saves you time, you reinvest that time into the next small improvement. Compounding, not a grand strategy. The question I try to ask myself daily is simple: how can I do my job a bit better today than I did it yesterday? Not more. Not faster. Better. I also don't run ten parallel AI workflows across different projects at the same time because someone told me that's what a serious AI-software engineer does now. If I have multiple projects going on, I only focus on one project at a time. That's the amount of mental space I have right now, and I've stopped treating that as a shortcoming. My impact shifted from writing code to making my team better The time that used to go into implementation didn't disappear, it moved upstream. I now sp

I've merged plenty of AI-generated code that was genuinely fine. I've also caught myself almost merging code that looked fine and wasn't, because it read like something a competent person wrote and my brain filled in the rest. Over the last year I've settled into a rough set of questions I run through before approving anything I didn't write line by line myself, generated or not. Here they are, in the order I actually ask them. 1. What problem is this code actually solving? It's easy to review whether code works and skip whether it solves the right thing. AI tends to answer the literal prompt, not the intent behind it. def get_active_users (): return db . query ( " SELECT * FROM users WHERE active = true " ) If "active" was supposed to mean "logged in within 30 days" and not a boolean flag that's rarely updated, this passes every test and still solves the wrong problem. Reviewer tip: Read the original ticket or request before reading the diff. Check the code against the intent, not just the literal ask. 2. Do I actually understand the implementation? Not "does it look reasonable," actually understand it, line by line, well enough to explain it to someone else. Reviewer tip: Try to explain the function out loud in one sentence per major step. If you get stuck anywhere, that's the part you haven't actually reviewed yet, just skimmed. 3. What assumptions is it making? Every implementation bakes in assumptions about the shape of the data, the order things happen in, or what "normal" looks like. function getLatestOrder ( orders ) { return orders [ orders . length - 1 ]; } This assumes orders is sorted chronologically and never empty. Neither assumption is stated anywhere. Reviewer tip: Ask "what does this assume about its inputs that isn't checked anywhere?" Write the answer down, literally, in the PR comment if it matters. 4. What happens with bad input? Bad input isn't an edge case, it's a certainty over a long enough timeline. def parse_age ( value ): return int ( val

This essay was written with Nathan E. Sanders, and originally appeared in The Guardian . OpenAI, and then Anthropic , were each formed by AI developers who feared unrestrained corporate AI development—specifically, that companies like Google and Meta would steer the technology towards deleterious, maybe even catastrophically unsafe, outcomes for society. Their founders proclaimed that their new labs, uniquely, could be trusted to develop the technology in humanity’s best interest. But each, in turn, were themselves co-opted by the same market incentives, themselves becoming corporate behemoths zealously guarding future investor value rather than the public interest...

Baruch Sadogursky and Patrick Debois discuss why coding agents fail due to bloated context windows and stuffed prompts. They explain practical context engineering fixes, including lazy-loaded skills, versioned context artifacts, externalized memory banks, and LLM-as-a-judge evals. Software architects & engineering leaders will learn how to turn raw markdown files into reliable agentic workflows. By Patrick Debois, Baruch Sadogursky

DeepSeek has announced price hikes for its V4 models.

The partnership is expanding beyond the initial market of Zagreb, Croatia to four additional European cities.

Expedia Group has open-sourced mockql-rs, a Rust CLI that fills @mock-annotated GraphQL fields with LLM-generated data at request time. It follows Airbnb's @generateMock in April and a GraphQL Foundation RFC opened in February. All three solve the same problem with different architectures, and two use the same directive name with incompatible semantics. By Steef-Jan Wiggers

From the multibrained octopus to the bitty brain organoid, we’re all just inference engines making flawed bets on the future. Here are the five insights your brain will need to survive.

One woman’s meth addiction was so bad, the only option left might have been brain surgery. Then a single session of noninvasive, focused ultrasound seemed to do what years of treatment could not.

For the past few months, I've been working on a project that translates online meetings in real time. At first, I thought the problem would mostly be about choosing the right speech recognition model and finding a good translation API. It turned out those were the easy parts. The real challenge was making everything feel instant. People don't wait for subtitles. If the translation appears two or three seconds after someone finishes speaking, the conversation has already moved on. Technically it works, but from a user's perspective, it feels broken. That completely changed how I approached the project. Instead of optimizing only for accuracy, I had to optimize the entire pipeline for latency: Capturing audio continuously Streaming audio to ASR Performing speech recognition incrementally Translating partial sentences Updating subtitles without flickering Handling corrections when the speech recognizer revised previous words Every stage might only take a few hundred milliseconds, but together they determine whether the experience feels "real-time." Another lesson surprised me even more. Translation quality isn't just about picking a better LLM. Spoken language is messy. People interrupt each other, change direction halfway through a sentence, use filler words, and rarely speak in complete grammatical sentences. A model that performs well on benchmarks can still struggle in a live conversation if the input arrives one fragment at a time. That forced me to rethink prompts, buffering strategies, and when to display or revise translated text. Building this project also gave me a new appreciation for streaming systems in general. Real-time applications are fundamentally different from batch processing. Instead of asking, "How accurate is the result?", you're constantly balancing three competing goals: Latency Stability Accuracy Improving one often makes another worse. I'm still learning every day, and there are plenty of problems left to solve. Over the next few weeks, I'd

Every generative-AI app I worked on hit the same wall. Pick a backend — OpenAI for images, Volcengine for video, Mureka for music — and before long the app is full of that provider's SDK quirks: its field names, its sync-vs-async polling loop, its error shapes. Then if want to swap one backend, or add a second one for failover, and it's a rewrite. mm-gateway is an open-source Python gateway that sits in front of that mess. One provider-neutral contract — over 13 backends : OpenAI · Google · xAI · DashScope · Volcengine · Flux · Stability · ElevenLabs · MiniMax · Mureka · ACE-Step · OpenRouter · UdioAPI The idea is simple: provider wire formats never appear in application code. Every request goes through a strict, modality-specific envelope — an ordered list of typed input parts plus provider-neutral parameters — and each backend adapter translates that to its native SDK or REST shape. For more information, visit https://github.com/sloth-os/mm-gateway

As someone who is constantly exploring ways to make AI applications faster and cheaper, I found...

1. Retomando: do Dockerfile mínimo a um Dockerfile de verdade Na segunda parte desta série, um Dockerfile de poucas linhas já foi suficiente para empacotar uma aplicação Python. Isso funciona, mas um Dockerfile escrito sem pensar em camadas e cache de build gera imagens maiores do que precisam ser e builds que demoram muito mais do que deveriam a cada mudança pequena no código. Este artigo aprofunda como o Docker constrói uma imagem por dentro, e como escrever um Dockerfile que tira proveito disso. 2. Como funcionam as camadas (layers) Cada instrução de um Dockerfile ( FROM , RUN , COPY , ADD ) que modifica o sistema de arquivos gera uma camada — um diff read-only armazenado separadamente e empilhado sobre as anteriores. A imagem final é simplesmente a soma de todas essas camadas, e o container em execução adiciona uma camada gravável no topo (union filesystem). Container (camada gravável) ────────────────────────── Camada 4: COPY . . Camada 3: RUN pip install -r requirements.txt Camada 2: COPY requirements.txt . Camada 1: FROM python:3.12-slim Duas consequências práticas importantes: Camadas são reaproveitadas entre imagens. Se duas imagens diferentes compartilham as mesmas primeiras instruções (por exemplo, a mesma FROM e o mesmo RUN apt-get install ), o Docker armazena essa camada uma única vez em disco, mesmo que várias imagens a usem. Camadas são cacheadas entre builds. Ao rodar docker build de novo, o Docker verifica cada instrução, na ordem: se a instrução e seus arquivos de entrada não mudaram desde o último build, ele reaproveita a camada já construída em vez de refazer o trabalho. Isso é a base de todo o próximo tópico. 3. Cache de build: ordenar o Dockerfile por frequência de mudança O cache de build é invalidado a partir do primeiro ponto de mudança : se a instrução N mudou (ou um arquivo que ela copia mudou), toda camada a partir de N é reconstruída — mesmo que as instruções seguintes sejam idênticas ao build anterior. Isso significa que a ordem das ins

Apple has reportedly trained a custom AI model for the China market alongside domestic tech giant Alibaba, a rare cross-border partnership that cuts across growing tensions between Beijing and Washington. The China-focused large language model was developed in partnership with Alibaba and trained with the company's support, Reuters reports, citing three unnamed people familiar with […]