cd/sources/dev-to· home sources Dev.to
cat /sources/dev-to.feed | wc -l → 11263

Dev.to

articles 11263 domain dev.to → page 471/564 feed RSS
07:11
2026-05-27
dev.to
artificial-intelligence

How Are Developers Actually Using AI At Work?

A developer named Sylwia, preparing for a discussion room at JSNation, is surveying how developers actually use AI in real work, beyond conference demos and viral stories. She shares an anecdote about…

06:44
2026-05-27
dev.to
artificial-intelligence

Practical AI Adoption in Test Automation

An engineer built OpenSecant, an AI-native resilient automation framework on Playwright, to address practical challenges in test automation. The framework uses a deterministic-first approach with loca…

06:16
2026-05-27
dev.to
artificial-intelligence

Building a Production MCP Server in Laravel

A Laravel developer has built a production-grade Model Context Protocol (MCP) server using HTTP+SSE transport, enabling external AI agents to discover, authenticate against, and invoke Laravel-based t…

05:37
2026-05-27
dev.to
machine-learning

The bf16 grad accumulator that killed our SDXL LoRA training

Photoroom's SDXL LoRA fine-tuning for a product photography model silently corrupted its adapter weights over six days due to a bf16 gradient accumulation issue. The custom training loop, forked from …

05:31
2026-05-27
dev.to
ai-tools

Mouse Unlock!—no password, just a secret click pattern

A developer created a mouse-based unlock system for their 10-year-old Linux laptop that has no fingerprint sensor or Face ID. The tool, built with Claude Code in approximately 150 lines of Python, use…

05:21
2026-05-27
dev.to
ai-tools

From Figma to production React, with AI in the loop

A developer has built shadcncraft, a design-to-code workflow that connects Figma directly to production React code using AI. The system uses four components—a Figma kit, a React registry, and two MCP …

05:18
2026-05-27
dev.to
ai-tools

AI 3D tools need product evals, not benchmark faith

A developer building AI-generated 3D tooling argues that public benchmarks should serve as lead signals, not product truth, because a model that scores well on an OpenSCAD-style benchmark can still pr…

← prev page 471 / 564 next →