DEV Community

#evals

Posts

đź‘‹ Sign in for the ability to sort posts by relevant, latest, or top.
Showing my work: the four artifacts behind an AI judge for generated courses (part 1)

Showing my work: the four artifacts behind an AI judge for generated courses (part 1)

1
Comments
8 min read
Our AI forecast lost to 'nothing changes'. Here is the test we now run on every forecast

Our AI forecast lost to 'nothing changes'. Here is the test we now run on every forecast

Comments
6 min read
Hy4 Flash walkthrough is up: what 1/20 the cost and half the coding score actually looks like on screen

Hy4 Flash walkthrough is up: what 1/20 the cost and half the coding score actually looks like on screen

1
Comments
2 min read
Your AI Model Is Not a Dependency Pin: Build an Eval Gate for Model Churn

Your AI Model Is Not a Dependency Pin: Build an Eval Gate for Model Churn

2
Comments
8 min read
LLM-as-judge evals: writing the rubric, and the failure modes that lie to you

LLM-as-judge evals: writing the rubric, and the failure modes that lie to you

Comments
3 min read
Run your own bake-off before you pick a production model

Run your own bake-off before you pick a production model

Comments
4 min read
"Trust Me Bro" Benchmarks — How AI Labs Grade Their Own Homework

"Trust Me Bro" Benchmarks — How AI Labs Grade Their Own Homework

Comments
29 min read
Evals before vibes

Evals before vibes

Comments
9 min read
MCP structuredContent: Claude Code Drops Your Text Blocks

MCP structuredContent: Claude Code Drops Your Text Blocks

1
Comments 2
9 min read
Why an 88.64 leaderboard score needed the catch in the same sentence

Why an 88.64 leaderboard score needed the catch in the same sentence

Comments
1 min read
Home-made Evals Just Taste Better

Home-made Evals Just Taste Better

3
Comments
8 min read
The Eval Scorers You Actually Need for LLM Apps (and When to Use Each)

The Eval Scorers You Actually Need for LLM Apps (and When to Use Each)

1
Comments
5 min read
Catch prompt regressions in CI before your users do

Catch prompt regressions in CI before your users do

Comments
3 min read
Can an AI model hide how capable it really is?

Can an AI model hide how capable it really is?

4
Comments
10 min read
LLM Evals Start as Thirty Test Cases, Not a Platform

LLM Evals Start as Thirty Test Cases, Not a Platform

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