DEV Community

#llm

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
System Prompt Confidentiality Is Security by Obscurity

System Prompt Confidentiality Is Security by Obscurity

Comments
4 min read
LLM Agent Sandboxes Have Two Escape Surfaces. Only One Gets CVEs.

LLM Agent Sandboxes Have Two Escape Surfaces. Only One Gets CVEs.

Comments 1
5 min read
I banned the em dash from my AI's voice

I banned the em dash from my AI's voice

Comments
4 min read
No card ships until a blind judge passes it

No card ships until a blind judge passes it

Comments
4 min read
handoff: Give the Next AI Agent the Context It Actually Needs

handoff: Give the Next AI Agent the Context It Actually Needs

Comments
5 min read
LLM Load Testing Is Burning Your API Budget

LLM Load Testing Is Burning Your API Budget

Comments
7 min read
The Orchestrator's Deputy Has No Scope: Ambient Authority Is the Root Bug in Multi-Agent AI

The Orchestrator's Deputy Has No Scope: Ambient Authority Is the Root Bug in Multi-Agent AI

Comments 1
5 min read
Flash Onyx 2.3: I shipped a model release with zero new weights

Flash Onyx 2.3: I shipped a model release with zero new weights

1
Comments
6 min read
Frontier LLM prices didn't move for 5 months. In August, they moved three times, and one lab tripled its rate.

Frontier LLM prices didn't move for 5 months. In August, they moved three times, and one lab tripled its rate.

Comments
6 min read
From Prompt Engineering to AI Engineering

From Prompt Engineering to AI Engineering

Comments
12 min read
Anthropic Drops Claude Sonnet 4.5: Extended Thinking Hits 200K Context for Agentic Coding

Anthropic Drops Claude Sonnet 4.5: Extended Thinking Hits 200K Context for Agentic Coding

1
Comments
3 min read
Every automated job names the model it runs on

Every automated job names the model it runs on

Comments
4 min read
When Your Benchmark Finally Tells the Truth

When Your Benchmark Finally Tells the Truth

10
Comments 2
8 min read
Four Releases in Four Days: Why Model Selection Is Now a Full-Time Job

Four Releases in Four Days: Why Model Selection Is Now a Full-Time Job

Comments
2 min read
Run GLM-5.3 Locally: Real Quant Sizes, the llama.cpp Surprise, and the reasoning_effort Trap

Run GLM-5.3 Locally: Real Quant Sizes, the llama.cpp Surprise, and the reasoning_effort Trap

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