DEV Community

Morgan Zhou profile picture

Morgan Zhou

DevOps enthusiast automating all the things.

Location London, UK Joined Joined on 

Education

UC Berkeley

Work

Frontend Developer at InnovateTech

Hand Them a Lease, Not a Key

Hand Them a Lease, Not a Key

Comments
7 min read
Separate the Spec from the Scratch Pad

Separate the Spec from the Scratch Pad

Comments
8 min read
Move the Box, Keep the Contract

Move the Box, Keep the Contract

Comments
7 min read
Put a Meter in the Take-Home

Put a Meter in the Take-Home

Comments
7 min read
Make the Trace the Deliverable

Make the Trace the Deliverable

Comments
6 min read
Rerun the Take-Home on a Clean Machine

Rerun the Take-Home on a Clean Machine

Comments
9 min read
Send the Scorecard With the Take-Home

Send the Scorecard With the Take-Home

Comments
7 min read
Grade the Tool Call, Not the Writeup

Grade the Tool Call, Not the Writeup

Comments
7 min read
Fetch the Trace Id Before You Hire

Fetch the Trace Id Before You Hire

Comments
9 min read
Hand Them a Wrong Answer On Purpose

Hand Them a Wrong Answer On Purpose

Comments
8 min read
Swap the Model Before You Score

Swap the Model Before You Score

Comments
7 min read
Score the Client They Point at Your Server

Score the Client They Point at Your Server

Comments
7 min read
Grade the Ways It Breaks

Grade the Ways It Breaks

1
Comments
8 min read
Give Them a Prompt They Can Finish Without a Paid Key

Give Them a Prompt They Can Finish Without a Paid Key

Comments
8 min read
Put the Rubric in the Take-Home Zip

Put the Rubric in the Take-Home Zip

Comments
8 min read
Grade the Receipt, Not the Agent Demo

Grade the Receipt, Not the Agent Demo

Comments
5 min read
The Take-Home Starts With a Red Log

The Take-Home Starts With a Red Log

Comments
7 min read
Hand Every Candidate the Same Machine

Hand Every Candidate the Same Machine

Comments
9 min read
Grade the Loop Budget Before You Grade the PR

Grade the Loop Budget Before You Grade the PR

Comments
8 min read
Put a Dead End in the Take-Home

Put a Dead End in the Take-Home

Comments
7 min read
Don't Page on a Null Status

Don't Page on a Null Status

Comments
7 min read
Grade Replay Before You Grade Style

Grade Replay Before You Grade Style

Comments
7 min read
The Retry Ticket Is the Interview

The Retry Ticket Is the Interview

Comments
10 min read
A Take-Home Packet That Catches Invented Infrastructure

A Take-Home Packet That Catches Invented Infrastructure

Comments
9 min read
Grade the Free Model Like a Take-Home

Grade the Free Model Like a Take-Home

Comments
9 min read
Mutation Testing on a Budget: Let a Free AI Model Sabotage Your Code

Mutation Testing on a Budget: Let a Free AI Model Sabotage Your Code

Comments
4 min read
The Free Model Server Smoke Test: 10 Minutes to Decide If It's Production-Worthy

The Free Model Server Smoke Test: 10 Minutes to Decide If It's Production-Worthy

Comments
4 min read
Let a Free AI Server Be Your Second Pair of Eyes on Every PR

Let a Free AI Server Be Your Second Pair of Eyes on Every PR

Comments
6 min read
Don't Upgrade Your AI Plan Until You've Benchmarked the Free One

Don't Upgrade Your AI Plan Until You've Benchmarked the Free One

Comments
4 min read
Audit Your AI Reviewer's Drift with a Nightly Control Run

Audit Your AI Reviewer's Drift with a Nightly Control Run

Comments
5 min read
A Free AI Server Is a Job Candidate. Interview It Like One.

A Free AI Server Is a Job Candidate. Interview It Like One.

Comments
5 min read
The Take-Home Test I Give Every AI Code Reviewer

The Take-Home Test I Give Every AI Code Reviewer

Comments
4 min read
Free AI Pull Request Reviews: A 20-Minute Setup

Free AI Pull Request Reviews: A 20-Minute Setup

Comments
3 min read
Stop Picking the Model First. Pick Where It Runs.

Stop Picking the Model First. Pick Where It Runs.

Comments
4 min read
Free, Self-Hosted, or Paid: Score Your Constraints First

Free, Self-Hosted, or Paid: Score Your Constraints First

Comments
5 min read
Free AI Coding Access vs. Self-Hosting: Pick Your Failure Mode

Free AI Coding Access vs. Self-Hosting: Pick Your Failure Mode

Comments
5 min read
A New Open-Weight Coding Model Drops Every Week. Here's How I Smoke-Test Them Before Trusting a Single Line

A New Open-Weight Coding Model Drops Every Week. Here's How I Smoke-Test Them Before Trusting a Single Line

Comments
6 min read
Cheap Seats for Agent Evals: Scoring Free Coding Models Against a Task Suite You Own

Cheap Seats for Agent Evals: Scoring Free Coding Models Against a Task Suite You Own

Comments
3 min read
Aider vs Claude Code in 2026: Open Source Freedom or Anthropic's Polish?

Aider vs Claude Code in 2026: Open Source Freedom or Anthropic's Polish?

Comments
2 min read
Quarantine the Output: Running Model-Generated Code Without Handing It Your Laptop

Quarantine the Output: Running Model-Generated Code Without Handing It Your Laptop

Comments
5 min read
Your AI Agent Has Shell Access. Here's How I Test What It Can Actually Touch.

Your AI Agent Has Shell Access. Here's How I Test What It Can Actually Touch.

Comments 1
5 min read
A Repeatable Harness for Catching Regressions in AI-Generated Code

A Repeatable Harness for Catching Regressions in AI-Generated Code

Comments
5 min read
Stop Trusting Vibes: A Repeatable Harness for Testing LLM-Generated Code on a Free Server

Stop Trusting Vibes: A Repeatable Harness for Testing LLM-Generated Code on a Free Server

Comments
5 min read
loading...