Recherche, sujets (tags des flux RSS), favoris et filtres de contenu.
Sujets les plus fréquents des 3 derniers jours. Un article doit porter tous les sujets choisis.
175 articles conservés pour « programming »
Software development involves considerably more than writing and reviewing code. Teams must document processes, coordinate releases, test integrations, investigate production failures and keep technical knowledge accessible as projects evolve. These responsibilities often…
AI is changing how we build apps, but what if AI could decide which UI components to display while our app stays in control of the design? That’s the idea behind A2UI (Agent-to-UI) . In my latest article, I explain A2UI in simple terms, using a restaurant analogy, and explore:…
EF Core is the most popular way to work with a database in .NET. Azure SQL Database is the managed SQL Server you get in the cloud, with backups, patching, and high availability handled for you. Putting the two together is straightforward, but the cloud adds a few things you…
Why orthogonal projections eliminate catastrophic forgetting in autoregressive Transformers—and the mathematical reason behind the 62% downstream plasticity penalty. The Continual Learning Dilemma in LLMs Fine-tuning Large Language Models sequentially without retraining from…
I recently added Find & Replace to YuHex, my independently developed hex editor for Windows. But I wanted to do things a little differently. Why I Built It This Way I've never really liked how some editors handle replacement. Replacing data one match at a time, or using "Replace…
How Taocarts Built a Referral Layer That Turns Buyers Into Resellers A common growth loop in cross-border reselling is simple: a happy buyer becomes a seller. They already trust the supply chain, they know the catalog, and they have their own audience back home. The natural next…
Quick answer: An AI coding agent is only as good as the context it is given and the verification it is held to. The two failures that dominate real work are under-context (the agent guesses the API and invents fields) and over-trust (it reports "done" after a green local run,…
The bug report always reads the same way. The extension works. Then the user leaves it alone for a minute, comes back, clicks the toolbar icon, and nothing happens. Or the badge count is wrong. Or the auth token the popup needs has vanished. Reload the extension and everything…
Short answer: don't trust them because they're green. AI writes tests that assert the code's current behavior — including the bugs. Earn the confidence by mutation testing: change the code on purpose and check the tests fail. The ones that stay green are lying to you. Last month…
GitHub Universe 2026 opened on October 28, 2026 in San Francisco with a message that is easy to miss if you only read the feature list: GitHub no longer thinks of Copilot as a coding assistant. It thinks of it as the delivery system itself — the thing that takes a task from an…
The Great Misunderstanding: Why We Fear AI Instead of Empowering It There is a pervasive narrative surrounding AI today, one that grips the public imagination with fear rather than possibility. It’s a common theme on social media and in tech circles: people refuse to see AI as…
You click Stop. The spinner disappears. The input unlocks. You give the agent a different task. Then the old tool finishes, and its result lands in the new conversation. Or worse: the old tool sends the message you thought you had stopped. These are different failures, and they…
The first month with an AI coding agent, I did what most people do: I got better at prompting. I learned to say "be minimal", to paste error logs, to ask for plans first. It helped a little. What actually changed things was unglamorous — I wrote the onboarding doc I would give a…
--- title: "Streaming LLM Tokens in Android with Jetpack Compose: Back-Pressure and the Frame Budget" published: true description: "Wire MediaPipe's LLM Inference API to Jetpack Compose using Kotlin Flow back-pressure, correct buffer sizing, and a 16ms recomposition budget that…
Knowing value categories in modern C++ is essential. Only when you know value categories, you can truly understand move semantics, and then you know how to write program efficiently. Let get straight into it! There are three primary categories. They are lvalue, xvalue and…
I'm building AgentGuard , an early-stage open-source project exploring permission controls for AI agents interacting with crypto wallets. The goal is to build a policy layer that can enforce: Transaction amount limits Daily spending limits Recipient allowlists Deterministic…
The audit had done its job. I knew where the generated audiobook didn’t match the manuscript. Now I had another problem. Most of the audio around those errors was already good. The pacing worked. The voices worked. The performances worked. Regenerating an entire chapter to fix…
Have you ever scanned a QR code that looked perfectly fine on a high-resolution screen, but failed miserably when printed? It happens all the time. Low contrast, missing quiet zones, or oversized logos can easily break the error correction data inside a QR code, rendering it…
Most businesses don't start looking for custom software because they want another dashboard. They start because something isn't working. Data is scattered across spreadsheets. Employees repeat the same tasks every day. An off-the-shelf tool handles 80% of the workflow, but the…
Ein praxisorientierter Leitfaden zu Datenschutz, Datenflüssen und Compliance-Nachweisen im Entwicklungsprozess. Datenschutz wird in vielen Softwareprojekten erst dann ausführlich behandelt, wenn ein Produkt kurz vor der Veröffentlichung steht oder eine Datenschutzprüfung…
AI collaboration tools are changing how teams plan projects, share information, and get work done. But I think there's an important distinction that often gets missed: helping people communicate faster is not the same as helping a team deliver work faster. A tool that summarizes…
Hello! If you saw my last post about SEA OS , you probably know that I had started building a kernel and a bootloader. Everything was going pretty well. I could build the project, boot it in QEMU, and see my kernel running. I was slowly getting closer to the things I actually…
I run a one-person R&D operation that ships industrial sensing products — analogue boards, microcontroller firmware, PC tools. Most of the typing is done by AI coding agents (mainly Claude Code). They are fast, and for a year they failed in the same four ways: They extended the…
If you are preparing for a Salesforce developer job, you probably know the feeling. You read the notes, watch the videos, and then freeze when an interviewer asks you a real question. I wanted a way to practice that feels like a real conversation, so I built a practice coach for…
This week's tooling landscape split cleanly between two themes: compliance infrastructure finally catching up to agentic AI workflows, and foundational APIs reaching the stability threshold where you can actually build on them. Zod and Java's Structured Concurrency both crossed…
Your Workflow Said Success and Nothing Happened: Verifying Side Effects in n8n This is part of my build-in-public series on running LLM workflows in n8n in production. Everything here is from real templates I sell and support — the mistakes are mine, the fixes are tested. Every…
A card holds a few bytes of opaque identifier. The number a reader shows you is a projection of those bytes — a base and a byte order chosen by whoever wrote the firmware — and nothing more. That is why the same credential can read as 04A2243A on the installer's laptop,…
If you use Cursor, you may have seen your default model flip to Grok. Cursor has been defaulting to Grok since SpaceX bought the company in August. The plans from 12 October include usage for Grok only. Other models like Claude are billed at API list price from the first…
Anyone who has given autonomous AI coding agents (Claude, Devin, Antigravity, or Copilot Workspace) free rein in a production repository knows the uneasy feeling: in one minute, the agent writes a brilliant algorithm; in the next, it touches an unrelated file, rewrites your…
Grounding an LLM judge: measure it on the model you run If you gate text with a one-word BLOCK/PASS LLM check, making every BLOCK quote the words it found can stop clean text from being refused: on the model we tested, clean comments went from 18 of 24 passing to 24 of 24. Test…
This week delivered two of the more practically meaningful cost reductions in recent memory—GPT-6.1 Terra cuts vision-reasoning spend in half with no migration pain, and the Decisions API finally makes structured classification a first-class API primitive. Layered on top: voice…
The Proof-Carrying Patch: Building a Verification Gate for AI-Generated Code A practical engineering tutorial for software developers and AI engineers By Akshat Raj | Software Engineering · AI Engineering · Developer Tooling Abstract AI coding agents can generate…
1/ Install Visual Studio Code 2/ Install Python and get the path where python is installed C:\Users\XXXXX\Python\pythoncore-3.15-64\python.exe python --version python path 3/ go to aistudio.google.com and use 300$ free credit. Create API Key 4/ Check the doumentation for api-key…
The oven is switched on, solar production drops and the car is charging. These actions involve different parts of the home, yet meet in the same electrical balance. An energy management system needs to know available power, respect device limits and decide which requests can…
Introduction Vim (the batch 24 post) is a legendary editor — and, by modern standards, a hard one to extend: its Vimscript configuration language and single-threaded design show their age. Neovim, forked from Vim in 2014, keeps Vim's modal editing and keybindings while…
Short answer: use a List<T> by default, and an array when the size never changes. But here's what most tutorials leave out: a List<T> is an array underneath. Watch a List grow List<int> numbers = new List<int>(); Console.WriteLine($"Count: {numbers.Count}, Capacity:…
The first time an AI assistant handed me a confident, beautifully formatted answer that turned out to be invented, it wasn't a disaster. It was a phone number. It looked right, it had the right area code, and it belonged to nobody. I caught it because I happened to know the real…
An AI feature can produce useful results in a demo and still encounter problems when connected to real data, users, and business workflows. For developers, moving from a prototype to a production application involves more than checking whether a model generates a relevant…
Construindo a Plinns: Novas features para devs (Integração com GitHub, Stacks e Importação de Artigos) Fala, comunidade! 👋 Há alguns dias compartilhei o início da Plinns (plinns.com), uma nova rede social com foco em conexões diretas e infraestrutura de alta performance. Tenho…
Your API is slow and Claude recommends you to add caching. It adds IMemoryCache to the shipment tracking API, and the p95 latency drops from 120 ms to 5 ms. Everyone is happy. Then they scale to two instances. A customer updates the delivery address, refreshes the page, and sees…
The rsync Algorithm Is Brilliant. Why Don't We See It Everywhere? Vadym Stupakov Vadym Stupakov Vadym Stupakov Follow Oct 19 The rsync Algorithm Is Brilliant. Why Don't We See It Everywhere? #cli #computerscience #networking #performance 1 reaction Add Comment 6 min read
TL;DR Instabase is built for large enterprises processing complex, multi-document packets, and it sells that way, through a demo and a custom quote, with no public pricing. A mid-market team that just needs invoices, statements, or forms extracted into usable fields often finds…
Root Causes of Silent Failures in FastAPI Exception Handling and Mitigation Patterns 1. Development Background and the Debugging Pitfall This tool was designed to traverse FastAPI endpoint functions (decorated with HTTP route methods) by compiling Python source code into an…
Introduction Express is the most popular web framework for Node.js, and it is deliberately tiny: a minimal layer on top of Node's built-in HTTP handling that adds routing and middleware, and little else. Where Django and Laravel include everything, Express includes almost…
Introduction Swift is Apple's language for its platforms: iPhone, iPad, Mac, Apple Watch, and Apple TV. Introduced in 2014 to replace the older Objective-C, it was designed to be safe, fast, and — unusually for a systems language — genuinely pleasant to read and write. Even if…
Introduction C# (pronounced "C sharp") is Microsoft's general-purpose language, and it is genuinely general: desktop apps, web APIs with ASP.NET, mobile apps, cloud services, and games built with Unity all run on it. If you learn one Microsoft-ecosystem language, this is the…
--> How Taocarts Turns a Storefront Order Into an Automated 1688 Purchase The part of a reverse-Taobao system people underestimate is the gap between "a customer paid on the overseas storefront" and "a supplier in China actually has the item reserved." That gap is where manual…
Module 5 of ML Zoomcamp hands you a frozen model and asks you to serve it. Sounds trivial until your probabilities are slightly off and nothing errors out. Here is the setup that worked for me, with the actual code. Verify the artifact first Before touching anything, check you…
Python's Integer Interning Trap a = 256 b = 256 print(a is b) # True x = 257 y = 257 print(x is y) # False (in REPL!) ⚡ The Small Integer Cache CPython allocates a fixed table of integer objects spanning [-5, 256] during interpreter initialization. Because integers are…
AI can write the code. Can you explain it? Learning to code has become a lot easier in some ways. A few years ago, if you wanted to build something, you would probably spend hours reading documentation, searching Stack Overflow, watching tutorials, and trying to understand why…