Topic
459 articles found
Claude writes. Codex reviews. Bugbot triages. Gemini sits on the council. I shipped v8.4.0 this week. Here's why the multi-model angle matters and what the harness actually does. The single-model problem Most coding agents (including the well-funded ones) run the whole SDLC on one model. The same we
A few weeks ago, I argued that AI is not a great equalizer — it's a great amplifier. It amplifies what developers already are, for better and for worse. Juniors with AI produce junior code at senior speed. Seniors with AI produce senior code at supernatural speed. This post is about where that ampli
Advisory CI gates are where good intentions go to die. A team adds a linter "in warning mode for now," and "for now" becomes forever. The violations scroll past in PR reviews, nobody cleans them, the gate never goes blocking. Six months later the warnings are archaeological noise. The pattern that b
AWS expanded its Savings Plans portfolio with Database Savings Plans, a spend-based discount model for managed database services that can cut costs by up to 35%. This is the first time the Savings Plans model has extended beyond compute, and it changes how DB teams can commit to long-term database s
The problem I am building Spanlens, an open-source LLM observability platform. Every call to OpenAI, Anthropic, or Gemini gets recorded with its model, latency, tokens, cost, and full request and response body. At low traffic on Supabase Postgres this was fine, but I could already see a few signs th
GitHub Copilot Has a New App. Here's What Changed for My Daily Workflow. GitHub launched a desktop-native Copilot app in technical preview this week. The key feature: agent sessions tied to a repo, issue, or PR — with a full session space that pauses, resumes, and can drive changes into a pull reque
Deep Research Agentic Systems are AI Agents designed to conduct multi-step research for complex tasks using dynamic reasoning, multi-hop information retrieval, and generate structured analytical reports. Sarang Kulkarni from Thoughtworks spoke at Arc of AI Conference 2026 on how to deploy multi-agen
curl is installed on roughly 30 billion devices. It's arguably the most scrutinised, most-fuzzed networking library on the planet. And right now, its creator is burning out. Not because curl is suddenly full of holes. Because AI-powered security research has reached a quality and volume that human m
The Problem We Were Actually Solving The treasure-hunt server receives 50 MB/s of dynamic map events—player moves, loot spawns, fog-of-war reveals—and must broadcast deltas to 100 k sockets without re-serializing the entire world every tick. config.yaml: world: width: 1024 height: 1024 chunk_size: 3
Let’s be honest for a second—Cursor AI is insanely powerful. You type a simple English prompt, hit a shortcut, and suddenly you have clean, working production code in front of you. It feels like you’re moving 10x faster and shipping features in minutes. But the uncomfortable truth is this: if you ar
Most outages aren't caused by bad code. They're caused by good code deployed in the wrong order. Senior developers don't rely on memory before a deploy. They run a checklist — every single time, even for a one-line change. Here's the exact checklist, and why each step exists. Pilots don't skip the p
the shift toward audit-first identity verification (https://go.caracomp.com/n/0527261218?src=devto) The recent news that the Australian Tax Office is procurement-ready for biometric liveness detection is a massive signal for developers in the computer vision and OSINT space. We are moving past the e
Apartment hunting is a stateful workflow. At any given moment, a user might have 15 listings in various stages: some saved, some inquired about, some with scheduled viewings, some attended, some applied to, some rejected. Each has associated data: emails, notes, documents, times, decisions pending.
When I looked at who was actually signing up to HomeScout in the early months, the timezone distribution surprised me. A meaningful chunk of users were browsing and setting up alerts from outside Ireland entirely. Netherlands, Germany, Australia, the US. People who needed a Dublin apartment but were
This is a submission for the GitHub Finish-Up-A-Thon Challenge I built and revived Hakozuna HZ5, an experimental page/run-first sidecar allocator prototype written in C. Hakozuna started as a personal allocator research project. Earlier versions focused on the HZ3/HZ4 allocator profiles: HZ3 / ACE-A
Building a Rental Aggregator When Daft.ie Already Exists When people find out I'm building a rental search platform for Dublin, the first question is usually some version of: "But why? Daft is already there." It's a fair question. Daft.ie has dominant market share, brand recognition, a large team, a
Data Normalization Across Dublin Rental Portals: How to Make Listings Comparable Dublin rental listings are fragmented even across the main portals. Daft.ie and Rent.ie use different structures, labels, price conventions, and quirks, which makes direct comparison harder than it should be. When I bui
If you run a freelance dev business or a small agency, you already know the silent margin-killer: Scope Creep. You sign a 30-page Statement of Work (SOW), start sprinting on the code, and three weeks later the client points to a vaguely worded bullet point from page 14 that suddenly means you owe th
Your AI agent has tool access. What it’s missing is the knowledge that makes those tools useful You’ve probably connected The post Why AI agents need a Context Lake appeared first on The New Stack.
The debate is loud. The takes are bad. Here’s the balanced version. - The coding world is split right now. Camp A: “Vibe coding isn’t real programming.” Camp B: “Syntax is dead. Just prompt it.” Both camps think they’re protecting the future of software. Both camps are partially right. Both camps ar
