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

Cheapest per token is not cheapest per finished job: what 2,400 runs show

A model with a 33% pass rate turned out to be the cheapest way to finish a task in a recent 10-model benchmark. The same benchmark shows why that result is less of a free lunch than it looks. Here is the metric behind it and how to compute it on your own prompts. The metric Cost per successful task = everything the model spent across every attempt (failed, retried, timed out) divided by the number of attempts that passed. Per-token price cannot see retries, malformed outputs or runs that never f

llmaibenchmark
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

I Let an AI Agent Prep My Workshop. Prep Took 20 Minutes.

Series: How to Run a Workshop People Remember (7/10) Workshop prep is the part nobody teaches and everybody dreads: read the run sheet, list the props, check print specs, build the sign-up and feedback forms, write the invite. It's all mechanical work that takes an evening — which is exactly the kind of work an AI agent should be doing while you do the human parts. The blocker was always that AI couldn't see your materials. It could draft you an email, but it couldn't open your activity folder,

aiworkshopproductivity
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

Keyword Research in Python: Ideas, Search Volume, Difficulty and Intent for ~$3 per 1,000 Keywords

Most keyword research comes down to one loop: start from a few seed terms, expand them into everything people actually search, then keep the ones with real volume and a difficulty you can win. Here's how to run that loop from code, paying per keyword instead of per month, plus what a real run returns and where the data falls short. Disclosure: we (SteadyScrape) built the tool below and get paid when it's used. Keyword Research Tool: Ideas, Volume, KD & Intent (steady-scrape/keyword-research-expl

seopythonmarketing
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

SilentMind: IA que censura e gera polêmica nas buscas

SilentMind: o modelo censurador que está bombando nas buscas e gerando polêmica Introdução Em 15 de outubro de 2024 a Tencent lançou o SilentMind, um modelo de IA que bloqueia automaticamente temas sensíveis e políticos. O diferencial? Quando o filtro dispara a resposta é cortada, substituída por “conteúdo não disponível” ou por mensagens genéricas. O anúncio disparou o Google Trends (↑ 420 % no Brasil, Portugal e México) e acendeu um debate acalorado sobre liberdade de expressão, c

anúnciosilentmindcensura
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

Tencent’s SilentMind AI Sparks Global Censorship Alarm

Tencent’s “SilentMind” AI Model Triggers a Wave of Censorship Concerns – What You Need to Know Introduction Tencent just launched SilentMind, an LLM that ships with a hard‑wired censorship layer blocking political dissent, Taiwan‑related terms, the 1989 Tiananmen incident, and a range of “social‑stability” content. Within 48 hours the model generated a 420 % surge in Google‑Trends searches for “SilentMind censorship,” and developers worldwide are scrambling to understand how the

tencentsilentmindcensorship
DEV CommunityRead original
ExternalSoftware Engineering
DEV Community

SilentMind de Tencent: descubre y elude sus filtros de censura

SilentMind de Tencent: ¿Qué filtra y cómo sortearlo? Introducción En menos de 48 h después del anuncio, SilentMind de Tencent se convirtió en el tema más buscado en Google Trends. ¿Por qué? Porque el modelo incorpora filtros internos que bloquean cualquier contenido que el gigante chino considere “sensibles” o “políticos”. Si trabajas con IA generativa, esa censura puede romper tu flujo de trabajo, tus pruebas y, en algunos casos, tu libertad de expresión. En este artículo te explic

silentmindcensuraiagoogletrends
DEV CommunityRead original
ExternalTechnology Trends
The Verge Tech

X says its updated web app is way faster

X is introducing a new version of its web app that it says is much faster than before, and it's now rolling out the updated app "one country at a time" for signed-in users. "Open a profile on X.com over a slow connection," principal engineer Yagiz Nizipli said in a blog post about the changes. "On the old web app, you'd stare at the X logo for more than 22 seconds before the first post appeared. On the new one, it shows up in 2.7." The post also goes into technical detail about how the updated a

The Verge TechRead 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]