DEV Community

#llm

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Why AI Agents Forget by Design

Why AI Agents Forget by Design

3
Comments
12 min read
I built Huiyu Pi — a self-hosted AI coding agent that starts at ~80 tokens.

I built Huiyu Pi — a self-hosted AI coding agent that starts at ~80 tokens.

Comments
1 min read
GrepSeek Trains a Search Agent to Use Shell Commands: GRPO-Trained Shell-Command Search

GrepSeek Trains a Search Agent to Use Shell Commands: GRPO-Trained Shell-Command Search

Comments
6 min read
Stop Your LLMs from Forgetting: How a 2016 String Algorithm Solves AI's Biggest Memory Loss Problem

Stop Your LLMs from Forgetting: How a 2016 String Algorithm Solves AI's Biggest Memory Loss Problem

11
Comments
8 min read
GPT-5.6 Is Real (a Codex Log Says So) — Everything Else Is Made Up

GPT-5.6 Is Real (a Codex Log Says So) — Everything Else Is Made Up

3
Comments
6 min read
Self-Hosted AI Risk Gate in 10 Minutes: Meet ITTE – Your Pre-Deploy Risk Brain with Self-Evolving Memory

Self-Hosted AI Risk Gate in 10 Minutes: Meet ITTE – Your Pre-Deploy Risk Brain with Self-Evolving Memory

Comments
1 min read
Intel Arc B580 for Local AI: 12 GB at $249, With a Software Tax

Intel Arc B580 for Local AI: 12 GB at $249, With a Software Tax

Comments
5 min read
From dynamic to adaptive: rewriting an agent's reasoning operation to its exact task at runtime

From dynamic to adaptive: rewriting an agent's reasoning operation to its exact task at runtime

Comments
2 min read
Anthropic Published a 31.5% Hijack Rate. Most Vendors Won't Even Show You a Number.

Anthropic Published a 31.5% Hijack Rate. Most Vendors Won't Even Show You a Number.

Comments
5 min read
JetBrains open-sources Mellum2 to challenge third-party API limitations

JetBrains open-sources Mellum2 to challenge third-party API limitations

Comments
5 min read
We ran an AI 'peer organization' (Claude + Codex + Gemini) for 7 weeks. Here is the operational record.

We ran an AI 'peer organization' (Claude + Codex + Gemini) for 7 weeks. Here is the operational record.

3
Comments 54
5 min read
Stop Caching LLM Responses. Cache the Thinking Instead.

Stop Caching LLM Responses. Cache the Thinking Instead.

1
Comments
2 min read
RAG Pipeline Chunking Strategies: Split Documents for Better Retrieval

RAG Pipeline Chunking Strategies: Split Documents for Better Retrieval

1
Comments 1
7 min read
How I Built an AI Document Pipeline That almost Never Hallucinates

How I Built an AI Document Pipeline That almost Never Hallucinates

Comments
2 min read
I got tired of AI agent explainers, so I built my own wiki

I got tired of AI agent explainers, so I built my own wiki

3
Comments
2 min read
đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.