DEV Community

#llm

Posts

👋 Sign in for the ability to sort posts by relevant, latest, or top.
Build a Token-Budgeted LLM Service on a Free Server: A Step-by-Step Tutorial

Build a Token-Budgeted LLM Service on a Free Server: A Step-by-Step Tutorial

Comments
5 min read
从 Demo 到生产:那些真正让 AI Agent 敢上线的护栏

从 Demo 到生产:那些真正让 AI Agent 敢上线的护栏

Comments
2 min read
Your Eval Set Is Rotting: A Theme-Fingerprint Drift Check

Your Eval Set Is Rotting: A Theme-Fingerprint Drift Check

1
Comments
4 min read
The Semantic Cache That Made a Free LLM Quota Feel Infinite

The Semantic Cache That Made a Free LLM Quota Feel Infinite

Comments
8 min read
MCP stdio Protocol's 3 Hidden Traps: When All Unit Tests Pass but the MCP Server Won't Respond

MCP stdio Protocol's 3 Hidden Traps: When All Unit Tests Pass but the MCP Server Won't Respond

Comments
13 min read
Microsoft archived PyRIT (Mar 2026) - what LLM red-teamers should use instead

Microsoft archived PyRIT (Mar 2026) - what LLM red-teamers should use instead

Comments
2 min read
Building a Self-Correcting AI Agent with Reflection Loops in Python

Building a Self-Correcting AI Agent with Reflection Loops in Python

Comments
5 min read
LLM Pricing This Week: DeepSeek Quietly Drops a Vision Model

LLM Pricing This Week: DeepSeek Quietly Drops a Vision Model

Comments
3 min read
G-CARL: Checklist-Aligned Reward Learning for Grounded Medical AI Agents

G-CARL: Checklist-Aligned Reward Learning for Grounded Medical AI Agents

Comments
5 min read
Causal Lineage and Session Replay with ZizkaDB

Causal Lineage and Session Replay with ZizkaDB

Comments
5 min read
I Tested 4 RAG Chunking Strategies Everyone Recommends. 2 Were Quietly Broken.

I Tested 4 RAG Chunking Strategies Everyone Recommends. 2 Were Quietly Broken.

Comments 2
6 min read
When Should You NOT Use an Agent?

When Should You NOT Use an Agent?

Comments
2 min read
Pi Coding Agent Review: Minimal, Hackable AI Coding CLI

Pi Coding Agent Review: Minimal, Hackable AI Coding CLI

Comments
16 min read
The AI writes the answer

The AI writes the answer

Comments
3 min read
We shipped two context-engineering features in one afternoon. We reverted them by dinner.

We shipped two context-engineering features in one afternoon. We reverted them by dinner.

3
Comments 1
4 min read
👋 Sign in for the ability to sort posts by relevant, latest, or top.