Development & Tech
Don't Let AI Count Blank Cells as Zero
A blank cell isn't zero. Before AI summarizes a spreadsheet, sort blanks into confirmed zero, unknown, and not applicable, and report the denominator and missing values with every ratio.
category
Development & Tech
Read mini classes on this topic together, starting with the latest lessons for context.
11
lessons
Recommended starts
Put stronger AI into reviewable workflows
As developer tools get stronger, teams need task framing, mobile approval, cost rules, uncertainty tests, and rollback design.
Mobile approval
Check evidence before approving coding agents on your phone
When you are away from a desktop, review diff, tests, artifacts, and rollback scope before approving.
Coding agents
Put human checkpoints first
Before handing code to AI, review the plan, small-scope execution, acceptance assumptions, and result.
Design handoff
Figma-to-code also needs acceptance boundaries
Before design becomes code, fix the diffs, states, and human review points.
Development & Tech
Before Letting AI Search for Faster Code, Write Down What "Better" Means
Tools like Google's AlphaEvolve generate many code variants and keep the top scorer, so anything the evaluator misses can be sacrificed. Write down baseline, score, hard limits, stop rules, and approver first.
Development & Tech
AI's First Version of a Code Change Is a Price Check; the Real Bill Is Review and Maintenance
AI made writing a first version cheap, but not reviewing or maintaining it. Treat it as a price check: merge only when someone can explain it without the AI chat and someone agrees to own it.
Development & Tech
Whether to Block an AI Crawler Depends on What It Came to Do
AI crawlers index for search, read pages live for a user, or take content for training. Each trades something different, so skip the single block-AI switch: set defaults by purpose and protect your most valuable pages.
Development & Tech
Before Merging AI-Written Code From Your Phone, Let Repository Rules Guard the Gate
Cursor's iPhone app starts AI coding agents, shows diffs, and can merge when they finish. Phones suit starting and tracking work; put what can't be merged into branch protection, not willpower.
Development & Tech
When a Third-Party Tool Is Breached, Your CRM Loss Depends on the Access You Granted
Attackers breached Klue, stole customers' Salesforce tokens, and pulled CRM data. You can't fix a vendor; you control how much it can read, whether unusual queries are seen, and how fast you revoke.
Development & Tech
An AI Container Scanner Draws the Line Itself: Mechanical Edits Can Be Automatic, Base Image Changes Need a Person
DockSec uses AI to explain Docker scan findings, but its auto-fix only makes mechanical Dockerfile edits, never base image picks, secrets, or compose files. Teams can adopt that line too.
Development & Tech
Whether Your Python Logs Are Good Enough Shows Up When You Replay the Last Incident
Good logs answer when, where, what happened, and with what data. Replay your last incident to see what's missing. Built-in logging can do it all; Loguru saves setup but logs variable values by default.
Development & Tech
With Copilot on Usage-Based Billing, Decide First Whether to Stop or Pay When Credits Run Out
Copilot bills by usage from June 2026: completions stay free, chat and agents cost, with no cheap-model fallback. Decide whether to stop or pay when credits run out, and who gets the priciest mode.
Development & Tech
Can a Coding Agent Take This Issue? Only If You Can Describe the First Diff
Coding agents can now go from issue to pull request, and AI can even approve the result. So the human checkpoint moves earlier: hand a task over only if you can describe its first independently reviewable diff.
Development & Tech
Writing "This Is False" Next to Bad Data Doesn't Stop AI From Believing It
Fine-tuning on documents that keep saying a claim is false can still teach the claim; models learn the negation mainly when it's inside the sentence. Store true/false status in fields or rewrite the data.