cd/sources/dev-to· home sources Dev.to
cat /sources/dev-to.feed | wc -l → 19195

Dev.to

articles 19195 domain dev.to → page 344/960 feed RSS
14:45
2026-08-09
dev.to
developer-tools

When AI Fails to Smash the Bug 🪲

A developer encountered a frustrating bug when deploying a local e-commerce shop through Cloudflare, where the site appeared broken despite working locally. After troubleshooting with AI assistance an…

14:34
2026-08-09
dev.to
developer-tools

Built an open-source MCP Conformance Scanner

A developer built the MCP Conformance Scanner, an open-source tool that helps developers validate their MCP servers before deployment. The scanner addresses common configuration and conformance issues…

14:29
2026-08-09
dev.to
large-language-models

Where Does RAG Actually Cost You Money? (Episode 6)

An engineer found that increasing the retrieval parameter top_k from 3 to 10 in a RAG system did not improve answer quality and sometimes made it worse, while also increasing token costs. By adding a …

14:05
2026-08-09
dev.to
artificial-intelligence

Keeping the LLM out of the verdict

A developer built an audit system that checks AI-generated fitness training plans against published sports-medicine guidelines, deliberately keeping LLMs out of the judgment path. The system splits in…

14:01
2026-08-09
dev.to
artificial-intelligence

MCP Went Stateless: What the 2026-07-28 Spec Actually Changes

The Model Context Protocol (MCP) became stateless at the protocol layer with the 2026-07-28 specification, removing the handshake and session ID header to enable horizontal scaling without sticky sess…

13:01
2026-08-09
dev.to
ai-safety

The Guardrail's Failure Mode Is Always "Allow"

Production AI guardrails are typically implemented as a second LLM or classifier call wrapped in a try/except that defaults to allowing content when the call fails, creating a silent fail-open safety …

13:00
2026-08-09
dev.to
ai-agents

Loaded Obeyed: Why Your Agent Quotes Rules It Doesn't Follow

A developer migrating from Claude Code to OpenCode found that agent rules were loaded but not followed, a state they call 'loaded but not load-bearing.' They propose a two-lane diagnostic model distin…

12:48
2026-08-09
dev.to
large-language-models

AI Agentic Memory for beginners.

A developer explains the fundamentals of AI agentic memory, describing how memory serves as context for large language models and how context windows impact response quality. The post highlights the c…

12:44
2026-08-09
dev.to
developer-tools

AmaliTech Apprenticeship Program (AAP) (AAP)

AmaliTech launched its Apprenticeship Program in Rwanda in November 2025, offering self-paced, in-person training in specializations such as fullstack development, Python backend and AI app developmen…

← prev page 344 / 960 next →