DEV Community

sichi chen profile picture

sichi chen

404 bio not found

Joined Joined on 
How do you score a pelican on a bicycle? Turning a viral SVG test into an evaluation you can audit

How do you score a pelican on a bicycle? Turning a viral SVG test into an evaluation you can audit

Comments
3 min read

Want to connect with sichi chen?

Create an account to connect with sichi chen. You can also sign in below to proceed if you already have an account.

Already have an account? Sign in
Stop trusting a 99% uptime claim: the LLM API relay failures we hit, and how we test stability

Stop trusting a 99% uptime claim: the LLM API relay failures we hit, and how we test stability

Comments
3 min read
Same model, different relay: a 5-metric checklist (with Python) before you pick an LLM API provider

Same model, different relay: a 5-metric checklist (with Python) before you pick an LLM API provider

Comments
2 min read
Before You Call an LLM Endpoint "Nerfed": A Small Statistics Checklist in Python

Before You Call an LLM Endpoint "Nerfed": A Small Statistics Checklist in Python

1
Comments 2
9 min read
Is the model actually getting dumber, or are we just reading tea leaves from single samples?

Is the model actually getting dumber, or are we just reading tea leaves from single samples?

Comments 5
2 min read
loading...