Medium

Antony

In

Level Up Coding

by

Antony

·

1d ago

Gemini 4 Carbon vs. Claude Fable 5.2: Which Should You Try First?

Compare coding benchmarks, API pricing, and the evidence each model can work with before you hand off a real ticket.

Gemini 4 Carbon vs. Claude Fable 5.2: Which Should You Try First?

In

Level Up Coding

by

Antony

·

4d ago

Could Grok Bot Save Me Time on My Morning AI News Brief?

Two official release notes show what I’d ask a scheduled news brief to find and what I’d still check myself.

Could Grok Bot Save Me Time on My Morning AI News Brief?

In

Level Up Coding

by

Antony

·

Nov 2

The Hardest Part of AI Coding Is Explaining What “Done” Means

A passing build can prove the implementation works while leaving the product question unanswered.

The Hardest Part of AI Coding Is Explaining What “Done” Means

In

CodeToDeploy

by

Antony

·

Oct 29

Which MacBook Pro Should You Buy in 2026? M6, M5 Pro, M5 Max, and the New Touch Display Explained

A practical guide to Apple’s two display families, the Singapore price gaps, and which upgrades are worth considering for the way you use a…

Which MacBook Pro Should You Buy in 2026? M6, M5 Pro, M5 Max, and the New Touch Display Explained

Antony

·

Oct 29

The Newest MacBook Chip Isn’t the Whole Upgrade Story

Apple’s latest launch makes a familiar buying shortcut harder to trust. For developers, the useful comparison starts with the work a…

The Newest MacBook Chip Isn’t the Whole Upgrade Story

In

Level Up Coding

by

Antony

·

Oct 28

When Your AI Agent Reads a Webpage, Who Is Giving the Instructions?

A document can be useful evidence without being allowed to tell your assistant what to do next.

When Your AI Agent Reads a Webpage, Who Is Giving the Instructions?

In

Level Up Coding

by

Antony

·

Oct 26

An AI Demo Is One Successful Run. A Product Has to Survive the Rest.

Repeated trials, realistic inputs, and visible failures tell you more than a polished example of what an agent can do.

An AI Demo Is One Successful Run. A Product Has to Survive the Rest.

In

CodeToDeploy

by

Antony

·

Oct 22

When AI Writes the Tests, Who Checks the Assumptions?

Generated code and generated tests can agree perfectly about the wrong behavior. Reviewing the expected result is where the useful work…

When AI Writes the Tests, Who Checks the Assumptions?

In

Level Up Coding

by

Antony

·

Oct 21

Your Codebase Is Now an Interface for AI Agents

Repository structure, executable checks, and documented constraints shape what an agent understands about your software.

Your Codebase Is Now an Interface for AI Agents

In

Level Up Coding

by

Antony

·

Oct 14

The AI Delegate Tax: Why Giving AI More Work Can Make Your Workflow Worse

The model bill is usually the smallest part of a bad AI workflow. The expensive part is the work that comes back unfinished, unverified, or…

The AI Delegate Tax: Why Giving AI More Work Can Make Your Workflow Worse

Antony

Antony

Software Engineer & Manager writing about AI automation, secure tech, investing, productivity systems, and self-improvement.

Following

Help

Status

About

Careers

Press

Blog

Store

Privacy

Rules

Terms

Text to speech