Skip to content
360SOFTY
HomeInsights

Insights

Ideas for building
better software.

Engineering perspectives on architecture, product development, AI, and the everyday decisions behind useful software.

From the engineering desk

Browse practical articles or follow the latest technology updates.

Subscribe via RSS →

Curated links from external sources — not 360Softy original articles.

ExternalSoftware Engineering
DEV Community

An AI wrote a working operating-system kernel from scratch in 38 minutes

An Anthropic model built a bootable operating-system kernel from an empty folder in roughly thirty-eight minutes of compute time, across about two hundred unassisted back-and-forth turns. The kernel boots inside an emulator and passes its own built-in tests — a task normally requiring months of specialist work. What: A blow-by-blow log shows one of the now-suspended models building bootable low-level systems code from an empty folder -- the kind of feat that made regulators nervous. When: 2026-0

codingagentsanthropiccapabilities
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

Our AI autopilot merged nothing for 3 days — the culprit was a review 'freshness' check

At Codens, we run our development on AI agents around the clock: a PRD becomes tasks, agents implement them and open PRs, an AI reviewer approves, and approved PRs merge automatically. Humans mostly read the reviewed output and the notifications. That was the idea — until one morning the dashboard showed zero completed tasks for three days straight. "Autopilot is down." Except it wasn't. Every component checked out healthy. Plans were being generated. Tasks were being submitted. Agents were open

aidevopsgithub
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

Sprint 7 closed: Engineering Agents

Sprint 7 is closed. The goal was specialized engineering agents — architecture, code review, documentation, debugging, and onboarding — built on skills, memory, and MCP. Agents gather evidence and return structured briefings; they do not call external LLMs. @meronq/agents — types, registry, context gathering, runner Five built-in agents — architecture, code_review, documentation, debugging, onboarding MCP + CLI — agent_list, agent_run, meronq agent run Handshake — available_agents[] for disc

devtoolsaiopensource
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

Stripped the best feature out of Google Antigravity and made it open source

Google Antigravity has this cool feature that bridges your IDE to a live browser tab using chrome extension where the AI model has access to live UI debugging capabilities. Built that exact visual debugging capability from scratch as an open-source tool, drop the paywall, works with all IDEs (VS code, Cursor, Warp, you name it) Checkout: https://github.com/DharuNamikaze/Gravity-lite https://www.npmjs.com/package/gravity-lite the writeup - https://dharun-builds.vercel.app/blog/gravity

antigravityopensourceprogramming
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

Our service discovery caught its own failure and switched itself off

We had a three-replica cluster that kept disagreeing with itself. Background jobs ran two and three times over. The answer wasn't in the logs; it was in our own code. The peer-discovery routine had a catch block, and the comment inside it said, more or less, "multicast failed, discovery disabled." Our service discovery was catching its own failure and quietly turning itself off, and it had been doing exactly that in production from day one. This is the story of why a clustering protocol that's c

authdistributedsystemsclustering
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

Importar usuários sem redefinição de senha

Todo guia de migração de identidade acaba chegando ao mesmo parágrafo, e ele soa sempre um pouco constrangido: "os usuários precisarão redefinir suas senhas." Isso é tratado como uma lei da natureza. Não é. É uma escolha, quase sempre imposta por uma ferramenta que não quis fazer o mais difícil. O mais difícil é verificar no lugar os hashes de senha existentes dos seus usuários, para que, depois da virada, eles entrem com exatamente as mesmas credenciais que tinham antes e nunca percebam que alg

migrationpasswordsbcrypt
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

Importar usuarios sin restablecer la contraseña

Toda guía de migración de identidad acaba llegando al mismo párrafo, y siempre suena un poco a disculpa: "los usuarios tendrán que restablecer su contraseña." Se trata como si fuera una ley de la naturaleza. No lo es. Es una decisión, casi siempre impuesta por una herramienta que no quiso hacer lo más difícil. Lo más difícil es verificar en su sitio los hashes de contraseña existentes de tus usuarios, para que tras el cambio inicien sesión con exactamente las mismas credenciales que tenían antes

migrationpasswordsbcrypt
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

Your RAG Pipeline Hallucinates Because It Never Checks Its Own Work

Your team ships a documentation chatbot. It retrieves chunks, stuffs them into a prompt, and generates an answer. Demo day goes great. Then a customer asks "what's the rate limit for the batch API?" and the bot confidently answers "10,000 requests per minute" — citing a doc about a completely different API. Nobody catches it because the answer sounds plausible. This is the core failure mode of naive RAG: the retriever returns something, the generator uses it, and nobody checks whether the retrie

aiprogramming
DEV CommunityRead original

Let’s start with a conversation

Tell us what you’re working on.

An idea, a challenge, or a system that needs to work better. We’ll help you understand the next step.

Discuss your project

Prefer email? [email protected]