DEV Community

# llm

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
CDD Paper: Context-Driven Decomposition for RAG Knowledge Conflict

CDD Paper: Context-Driven Decomposition for RAG Knowledge Conflict

Comments
9 min read
My AI Remembers Its Mistakes. Permanently. Here's the Engineering.

My AI Remembers Its Mistakes. Permanently. Here's the Engineering.

Comments
13 min read
We scanned 50+ MCP servers and found HIGH-severity bugs in Atlassian, GitHub, Cloudflare, and Microsoft — here's what we learned

We scanned 50+ MCP servers and found HIGH-severity bugs in Atlassian, GitHub, Cloudflare, and Microsoft — here's what we learned

1
Comments 1
4 min read
Why I Built SemanticGuard

Why I Built SemanticGuard

Comments
6 min read
My agent swarm had a productive night. My pipeline lied about it.

My agent swarm had a productive night. My pipeline lied about it.

1
Comments 1
5 min read
Why your local LLM aces benchmarks but fails real terminal tasks

Why your local LLM aces benchmarks but fails real terminal tasks

2
Comments 1
5 min read
Why AI Hallucinates

Why AI Hallucinates

1
Comments
3 min read
Meet pixserp — One Drop-in API for Web, News, Places, Flights, Hotels, YouTube and Anything Else on the Live Web

Meet pixserp — One Drop-in API for Web, News, Places, Flights, Hotels, YouTube and Anything Else on the Live Web

Comments
6 min read
MoE Architectures Keep Solving the Wrong Problem

MoE Architectures Keep Solving the Wrong Problem

Comments
3 min read
Why prompt engineering fails for tone control — and how steering vectors fix it

Why prompt engineering fails for tone control — and how steering vectors fix it

1
Comments
5 min read
Why Enterprise AI Systems Need Rollback Strategies Like Traditional Software

Why Enterprise AI Systems Need Rollback Strategies Like Traditional Software

Comments
3 min read
Raw HTML is where LLM context goes to die

Raw HTML is where LLM context goes to die

1
Comments
5 min read
10 Models Tested: From 81.6% to 10%. The Free Tier is a Full-On Gamble.

10 Models Tested: From 81.6% to 10%. The Free Tier is a Full-On Gamble.

Comments
4 min read
We Asked 10 LLMs to Write Efficient Code. Only 4 Got Better.

We Asked 10 LLMs to Write Efficient Code. Only 4 Got Better.

Comments
5 min read
MCP is quietly commoditizing data+model SaaS moats — the structural case

MCP is quietly commoditizing data+model SaaS moats — the structural case

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