DEV Community

Brian Jin profile picture

Brian Jin

Open-source maintainer building Judgment Pack, a testable specification and runtime for AI-agent decisions. Former enterprise data and software architect.

Location Toronto, Ontario, Canada Joined Joined on  Personal website https://judgmentpack.org/ github website

Work

Maintainer, Judgment Pack Specification

I stopped printing the thresholds in the policy. The AI wrote the same tests anyway.
kikashy profile

I stopped printing the thresholds in the policy. The AI wrote the same tests anyway.

#ai #testing #opensource
Comments
9 min read

Want to connect with Brian Jin?

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

Create Account
Already have an account? Sign in
When Agents Can Act on Real Devices, Governance Moves Beyond the Prompt
kikashy profile

When Agents Can Act on Real Devices, Governance Moves Beyond the Prompt

#ai #agents #security #architecture
Comments 4
6 min read
The Receipt Was Valid. The Policy Was Retired.
kikashy profile

The Receipt Was Valid. The Policy Was Retired.

#ai #security #testing #opensource
1
Comments 2
9 min read
Five Adversarial Reviews Told Me My Study Was Measuring a Fiction
kikashy profile

Five Adversarial Reviews Told Me My Study Was Measuring a Fiction

#ai #opensource #testing #architecture
1
Comments 1
11 min read
When Agents Can Run Anywhere, Where Should Organizational Judgment Live?
kikashy profile

When Agents Can Run Anywhere, Where Should Organizational Judgment Live?

#ai #opensource #agents #architecture
6
Comments 2
5 min read
The Receipt Was Cryptographically Valid. The Action Didn't Match the Judgment.
kikashy profile

The Receipt Was Cryptographically Valid. The Action Didn't Match the Judgment.

#ai #security #testing #opensource
1
Comments 1
11 min read
I Broke My AI Decision System 20 Ways to See Which Layer Would Catch What
kikashy profile

I Broke My AI Decision System 20 Ways to See Which Layer Would Catch What

#ai #agents #testing #opensource
Comments
8 min read
Where Does Judgment End and Runtime Policy Begin?
kikashy profile

Where Does Judgment End and Runtime Policy Begin?

#ai #opensource #agents #architecture
3
Comments 1
4 min read
The Agent Stack Is Filling In. The Judgment Layer Is Still an Open Problem
kikashy profile

The Agent Stack Is Filling In. The Judgment Layer Is Still an Open Problem

#ai #opensource #agents #architecture
Comments
5 min read
I Asked an AI to Author the Same Policy Tests 50 Times. It Hit Every Boundary in 49 Valid Runs.
kikashy profile

I Asked an AI to Author the Same Policy Tests 50 Times. It Hit Every Boundary in 49 Valid Runs.

#showdev #ai #opensource #machinelearning
13
Comments 7
10 min read
What I Learned Trying to Make AI-Agent Decisions Testable
kikashy profile

What I Learned Trying to Make AI-Agent Decisions Testable

#showdev #ai #opensource
Comments 7
6 min read
loading...