DEV Community

#llm

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Portable AI on a USB Stick

Portable AI on a USB Stick

Comments
1 min read
How llm-d Prefix-Cache Routing Made Qwen 7B on EKS 2.3x Faster

How llm-d Prefix-Cache Routing Made Qwen 7B on EKS 2.3x Faster

Comments
7 min read
Skills as Sub-Agents: Orchestrating Complex work with Claude Skills

Skills as Sub-Agents: Orchestrating Complex work with Claude Skills

1
Comments
7 min read
Auto-Compact Ate My Constraints: A Post-Mortem

Auto-Compact Ate My Constraints: A Post-Mortem

2
Comments 2
3 min read
How I Built a No-Execution LLM Eval Judge (75% Accuracy, No API Calls)

How I Built a No-Execution LLM Eval Judge (75% Accuracy, No API Calls)

1
Comments
2 min read
How Do AI Agents Remember Things Between Conversations?

How Do AI Agents Remember Things Between Conversations?

Comments
4 min read
D+T2 names who enters; budget names who gets seen

D+T2 names who enters; budget names who gets seen

3
Comments 16
6 min read
AI Harnesses Are Just Middleware, and Middleware Trust Bugs Are Older Than Your Career

AI Harnesses Are Just Middleware, and Middleware Trust Bugs Are Older Than Your Career

1
Comments 2
3 min read
Bringing Notes, WeChat Reading, and Zhihu into Obsidian: My LLM-Wiki Knowledge Hub

Bringing Notes, WeChat Reading, and Zhihu into Obsidian: My LLM-Wiki Knowledge Hub

1
Comments
7 min read
We Asked 10 LLMs to Recommend Brands. They Gave Us 29 Different Answers.

We Asked 10 LLMs to Recommend Brands. They Gave Us 29 Different Answers.

1
Comments
3 min read
Calling a Model Is Easy. Running an AI System Is Not

Calling a Model Is Easy. Running an AI System Is Not

Comments
7 min read
Your Scaffold Will Be Gamed

Your Scaffold Will Be Gamed

1
Comments
6 min read
What Is Agentic AI? And Why Oversight Has to Change

What Is Agentic AI? And Why Oversight Has to Change

1
Comments 1
8 min read
LLMs on Consumer Hardware — Part 3: A Multi-Agent Setup with Hermes — Delegation and Early Failure Modes

LLMs on Consumer Hardware — Part 3: A Multi-Agent Setup with Hermes — Delegation and Early Failure Modes

Comments
5 min read
Anthropic admits Claude breached three live corporate networks during safety tests

Anthropic admits Claude breached three live corporate networks during safety tests

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