DEV Community

#llm

Posts

👋 Sign in for the ability to sort posts by relevant, latest, or top.
One TPU Chip, Eight Agents: Serving Small Agent Workloads with Raw JAX

One TPU Chip, Eight Agents: Serving Small Agent Workloads with Raw JAX

5
Comments 2
15 min read
Testing Non-Deterministic LLM Pipelines in CI: A Contract-Based Approach

Testing Non-Deterministic LLM Pipelines in CI: A Contract-Based Approach

2
Comments 1
4 min read
Why Kimi K3 Still Can't Do What Einstein Did

RAG surfaces echoes, but misses paradigm shifts

Why Kimi K3 Still Can't Do What Einstein Did

22
Comments 12
4 min read
Your cache_read_input_tokens is zero. Here is what silently did it.

Your cache_read_input_tokens is zero. Here is what silently did it.

Comments
5 min read
I gave the same fabricated answer to RAGAS and DeepEval. One scored it 0.0. The other scored it 1.0

I gave the same fabricated answer to RAGAS and DeepEval. One scored it 0.0. The other scored it 1.0

Comments
6 min read
LOCKS — per-page SVD cuts KV cache reads 10 at 1M context

LOCKS — per-page SVD cuts KV cache reads 10 at 1M context

Comments
4 min read
Close your editor before heavy jobs? The heavy job lives inside my editor

Close your editor before heavy jobs? The heavy job lives inside my editor

Comments
4 min read
How I Ran Gemma 4 26B on M-Series Mac: 2GB RAM, 1.8 tok/s

How I Ran Gemma 4 26B on M-Series Mac: 2GB RAM, 1.8 tok/s

Comments
9 min read
The Biggest AI Stories Weren’t Features. They Were Dependency

The Biggest AI Stories Weren’t Features. They Were Dependency

Comments
12 min read
Four Models Cited My Numbers Perfectly. One Still Misread Them.

Four Models Cited My Numbers Perfectly. One Still Misread Them.

Comments
6 min read
OpenEval: Why LLM Evaluation Needs a Standard Format

OpenEval: Why LLM Evaluation Needs a Standard Format

Comments
1 min read
AI Agent Security Audit: From MCP Penetration Testing to LLM Vulnerability Assessment

AI Agent Security Audit: From MCP Penetration Testing to LLM Vulnerability Assessment

Comments
4 min read
Your AI Subagents Are Lying to You: 4 Silent Failure Modes

Your AI Subagents Are Lying to You: 4 Silent Failure Modes

1
Comments 3
4 min read
Loop Engineering Is Mostly Papering Over a Model That Won't Converge

Loop Engineering Is Mostly Papering Over a Model That Won't Converge

1
Comments
4 min read
Why does parsing scientific papers for RAG still break on equations and tables?

Why does parsing scientific papers for RAG still break on equations and tables?

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