DEV Community

#llm

Posts

👋 Sign in for the ability to sort posts by relevant, latest, or top.
Customer Support Chatbot Runtime Economics: LLM API Compatibility Beyond Token Price

Customer Support Chatbot Runtime Economics: LLM API Compatibility Beyond Token Price

Comments
6 min read
What a Neural Network Actually Computes: From Token IDs to Matrix Multiplication

What a Neural Network Actually Computes: From Token IDs to Matrix Multiplication

Comments
4 min read
Your Prompt Engineering Is Not the Bottleneck Anymore

Your Prompt Engineering Is Not the Bottleneck Anymore

Comments 1
4 min read
LLMs Have Learned All Open Code on GitHub - But Missed the Most Important Part.

LLMs Have Learned All Open Code on GitHub - But Missed the Most Important Part.

Comments
3 min read
117 Ghost Errors: Anatomy of a Flaky AI Agent

117 Ghost Errors: Anatomy of a Flaky AI Agent

1
Comments 1
10 min read
Where to cut a prompt so the cache actually hits

Where to cut a prompt so the cache actually hits

Comments
4 min read
Stop Trusting Vibes: A Reproducible Harness for Comparing AI Coding Models on Your Own Codebase

Stop Trusting Vibes: A Reproducible Harness for Comparing AI Coding Models on Your Own Codebase

Comments
5 min read
5 Undocumented Rules for Gemini Structured Output, Measured in Production

5 Undocumented Rules for Gemini Structured Output, Measured in Production

10
Comments 2
8 min read
No, the LLM Doesn't Get to Approve Your Refund

No, the LLM Doesn't Get to Approve Your Refund

Comments 2
6 min read
Is the future of AI local?

Is the future of AI local?

Comments
4 min read
Self-Hosting a Terminal AI Coding Assistant (Ollama + Tailscale + Qwen Code)

Self-Hosting a Terminal AI Coding Assistant (Ollama + Tailscale + Qwen Code)

Comments
5 min read
Claude Code Subagents: Setup, Config, and When to Use Them

Claude Code Subagents: Setup, Config, and When to Use Them

Comments
11 min read
My Tool-Calling Loop Worked Fine, Until Compliance Wanted a Second Model to Check It

My Tool-Calling Loop Worked Fine, Until Compliance Wanted a Second Model to Check It

2
Comments 1
4 min read
Score Coding Models With a 60-Line Harness Before You Spend a Cent

Score Coding Models With a 60-Line Harness Before You Spend a Cent

Comments
5 min read
What I learned trying to benchmark local LLMs honestly

What I learned trying to benchmark local LLMs honestly

Comments 2
4 min read
👋 Sign in for the ability to sort posts by relevant, latest, or top.