DEV Community

#llm

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Model Poisoning: The Hidden Risk in Supply Chain AI

Model Poisoning: The Hidden Risk in Supply Chain AI

Comments
4 min read
Multi-agent fleets burn ~15x the tokens — here's the budget layer the playbooks skip

Multi-agent fleets burn ~15x the tokens — here's the budget layer the playbooks skip

1
Comments
5 min read
We added synthetic data to our eval set. The pass rate rose, and so did our production incidents.

We added synthetic data to our eval set. The pass rate rose, and so did our production incidents.

Comments 2
5 min read
Capping VLM spend per CV researcher: hierarchical budgets in practice

Capping VLM spend per CV researcher: hierarchical budgets in practice

1
Comments 2
4 min read
An AI on our team faked a tool result. Here's the detector we shipped.

An AI on our team faked a tool result. Here's the detector we shipped.

Comments 13
8 min read
Token Saving, and Caveman

Token Saving, and Caveman

Comments
7 min read
GLM 5.2 isn't free: not even my US$4,000 Spark can run it

GLM 5.2 isn't free: not even my US$4,000 Spark can run it

3
Comments
6 min read
CacheWeaver Reorders RAG Evidence for Prefix-Cache Reuse: Prefix-Cache-Aware Evidence Reordering

CacheWeaver Reorders RAG Evidence for Prefix-Cache Reuse: Prefix-Cache-Aware Evidence Reordering

1
Comments 3
7 min read
I Built an AI Agent That Gets Curious On Its Own

I Built an AI Agent That Gets Curious On Its Own

3
Comments 6
4 min read
My AI Wrote Code That Passed Every Test and Was Still Wrong

My AI Wrote Code That Passed Every Test and Was Still Wrong

Comments 2
4 min read
I tested 5 LLMs for prompt-injection leaks. Same code, 0% to 90%.

I tested 5 LLMs for prompt-injection leaks. Same code, 0% to 90%.

Comments
3 min read
I don't trust the LLM to classify my email. So I don't let it.

LLM as feature scorer, code as decider

I don't trust the LLM to classify my email. So I don't let it.

15
Comments 25
5 min read
The stale context problem: why your AI doesn't know what time it is

The stale context problem: why your AI doesn't know what time it is

2
Comments 2
3 min read
Has anyone else seen prompt caching break because of UUIDs/timestamps near the front?

Has anyone else seen prompt caching break because of UUIDs/timestamps near the front?

1
Comments 2
1 min read
I Installed 50 AI Agent Skills Blindly. Here's the 3-Question Filter I Use Now.

I Installed 50 AI Agent Skills Blindly. Here's the 3-Question Filter I Use Now.

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