Jordan Liu

Jordan Liu

Sep 21

I Counted Drops as Wrongs. The Chart Was Theater.

#ai #python #testing #debugging

7 min read

Jordan Liu

Sep 20

I Subtracted the Queue. The Fast Model Wasn't.

#ai #testing #python #performance

6 min read

Jordan Liu

Sep 19

I Imported the Patch. That's When It Died.

#python #ai #testing #opensource

6 min read

Jordan Liu

Sep 18

I Shuffled the Judge Prompt. The Scores Should Have Held.

#ai #testing #python #agents

7 min read

Jordan Liu

Sep 17

I Scored Patches That Never Applied

#ai #testing #python #git

6 min read

Jordan Liu

Sep 16

I Scored Tests That Could Not Fail

#python #ai #testing #agents

7 min read

Jordan Liu

Sep 14

Free Inference Has a Tail. I Probe It Before I Ship.

#ai #agents #performance #testing

6 min read

Jordan Liu

Sep 13

I Scored Duplicate Tool Calls. Busy Was Not Progress.

#ai #testing #agents #python

7 min read

Jordan Liu

Sep 12

I Scored Retry Loops. Boundedness Was Optional.

#ai #python #testing #agents

1

8 min read

Jordan Liu

Sep 11

I Counted the Turns That Didn't Need a Model

#ai #testing #agents #python

8 min read

Jordan Liu

Sep 9

I Asked for One Line. I Scored the Blast Radius.

#python #testing #ai #git

7 min read

Jordan Liu

Sep 6

I Failed the Run for a Timezone Nobody Said

#ai #testing #agents #python

8 min read

Jordan Liu

Sep 5

Valid JSON Is Not a Grounded Tool Call

#ai #python #testing #agents

9 min read

Jordan Liu

Sep 4

The Free Endpoint Passed Chat. It Failed the Schema Fuzzer.

#ai #python #testing #json

6 min read

Jordan Liu

Sep 4

My Free Server Died at 3 AM. Here's the Debugging Trail.

#ai #serverless #webdev #tutorial

4 min read

Jordan Liu

Sep 3

I Stopped Scoring Completions. I Started Scoring Invariants.

#python #ai #testing #llm

8 min read

Jordan Liu

Sep 2

My Free Model's Confidence Score Was a Liar. Here's the Probe.

#ai #python #testing #llm

4 min read

Jordan Liu

Sep 1

Free Models and a Free Server: Building a Zero-Budget LLM Regression Harness

#ai #llm #testing #devtools

5 min read

Jordan Liu

Aug 31

Free AI Servers Hide Their Worst Behavior. This Probe Exposes It.

#ai #testing #python #productivity

4 min read

Jordan Liu

Aug 30

I Wrote a Token Budget Probe for a Free Model Server. The Numbers Matter.

#ai #testing #productivity #python

4 min read

Jordan Liu

Aug 29

I Timed a Free Model Server at 3 AM, Noon, and 9 PM. Peak Hours Are a Different Product.

#ai #testing #performance #opensource

4 min read

Jordan Liu

Aug 28

I Stress-Tested a Free Model Server. The Server Survived. My Retry Logic Didn't.

#ai #llm #python #testing

4 min read

Jordan Liu

Aug 26

I Benchmarked a Free AI Code Reviewer. It Caught 23 of 40 Bugs and Invented 6.

#ai #testing #productivity #opensource

4 min read

Jordan Liu

Aug 25

I Let a Free Model Write My Commit Messages. The Gate Was the Real Author.

#ai #python #git #productivity

5 min read

Jordan Liu

Aug 25

I Let a Free Model Audit My Error Messages. 22 Out of 47 Were Useless.

#python #ai #productivity #devtools

1

5 min read

Jordan Liu

Aug 24

I Built a CI Failure Digest on a Free Server. The Model Was the Easy Part.

#ai #python #devops #opensource

4 min read

Jordan Liu

Aug 23

I Logged Every Model Decision My Triage Agent Made. The Log Was the Real Product.

#ai #agents #python #webdev

5 min read

Jordan Liu

Aug 21

The Refactor Had No Tests. So I Asked a Free Model What "Correct" Meant.

#testing #refactoring #ai #python

4 min read

Jordan Liu

Aug 14

MiniMax H3 Is a Model, Not a Routing Decision: A Repeatable Gate

#ai #agents #python #opensource

5 min read

Jordan Liu

Aug 13

I Put My Agent's Model Routing Policy in Git. The Bill Dropped Before the Code Did.

#ai #llm #automation #python

2

4 min read

Jordan Liu

Aug 10

Your Agent's Permission Slip Belongs in Version Control, Not in a Prompt

#security #ai #agents #testing

7 min read

DEV Community

Badges

Writing Debut

Skills/Languages

Currently learning

Currently hacking on

Available for

I Counted Drops as Wrongs. The Chart Was Theater.

I Subtracted the Queue. The Fast Model Wasn't.

I Imported the Patch. That's When It Died.

I Shuffled the Judge Prompt. The Scores Should Have Held.

I Scored Patches That Never Applied

I Scored Tests That Could Not Fail

Free Inference Has a Tail. I Probe It Before I Ship.

I Scored Duplicate Tool Calls. Busy Was Not Progress.

I Scored Retry Loops. Boundedness Was Optional.

I Counted the Turns That Didn't Need a Model

I Asked for One Line. I Scored the Blast Radius.

I Failed the Run for a Timezone Nobody Said

Valid JSON Is Not a Grounded Tool Call

The Free Endpoint Passed Chat. It Failed the Schema Fuzzer.

My Free Server Died at 3 AM. Here's the Debugging Trail.

I Stopped Scoring Completions. I Started Scoring Invariants.

My Free Model's Confidence Score Was a Liar. Here's the Probe.

Free Models and a Free Server: Building a Zero-Budget LLM Regression Harness

Free AI Servers Hide Their Worst Behavior. This Probe Exposes It.

I Wrote a Token Budget Probe for a Free Model Server. The Numbers Matter.

I Timed a Free Model Server at 3 AM, Noon, and 9 PM. Peak Hours Are a Different Product.

I Stress-Tested a Free Model Server. The Server Survived. My Retry Logic Didn't.

I Benchmarked a Free AI Code Reviewer. It Caught 23 of 40 Bugs and Invented 6.

I Let a Free Model Write My Commit Messages. The Gate Was the Real Author.

I Let a Free Model Audit My Error Messages. 22 Out of 47 Were Useless.

I Built a CI Failure Digest on a Free Server. The Model Was the Easy Part.

I Logged Every Model Decision My Triage Agent Made. The Log Was the Real Product.

The Refactor Had No Tests. So I Asked a Free Model What "Correct" Meant.

MiniMax H3 Is a Model, Not a Routing Decision: A Repeatable Gate

I Put My Agent's Model Routing Policy in Git. The Bill Dropped Before the Code Did.

Your Agent's Permission Slip Belongs in Version Control, Not in a Prompt