AI & prompting: page 2

In “How to write a resume with AI using experience you can prove”

How do I stop AI from making up resume achievements?

Supply a verified evidence list and explicitly prohibit invented metrics, titles, skills and outcomes. Then audit the revision against that list yourself. Prompt instructions can reduce unwanted additions but cannot guarantee their absence. If a number or claim has no source, remove it or confirm it before submitting.

How to use AI for Excel and Google Sheets formulas

Name your spreadsheet app, define the columns and paste a small anonymized sample. State the result you expect and the rules for blanks, duplicates and errors. Ask for a formula plus a short explanation, run it in a copy, and compare it with a hand-calculated answer before extending it to your real rows.

In “How to use AI for Excel and Google Sheets formulas”

Can ChatGPT write Excel formulas without seeing my whole file?

Yes. Describe the column letters, headers, relevant rows, spreadsheet version and desired result, then paste a small anonymized example. Include a result you can calculate yourself. Test the suggested formula in your spreadsheet; a plausible explanation alone does not show that the formula references the right data.

In “How to use AI for Excel and Google Sheets formulas”

Why does an AI formula work in Excel but fail in Google Sheets?

The suggested formula may use a feature, function or reference style that differs between the apps, or your locale may use different separators. Tell the assistant the exact app and paste the error. Ask for a compatible formula using your sample, then check the result rather than changing several settings at once.

In “How to use AI for Excel and Google Sheets formulas”

How do I check that an AI spreadsheet total is correct?

Start with a few rows and add the relevant values by hand. Then change one contributing amount and one category to confirm that the formula responds as expected. Check the first and last included rows, blanks, numeric text and duplicate rules before extending the formula to the full dataset.

In “How to use AI for Excel and Google Sheets formulas”

Should I upload my customer spreadsheet to get formula help?

Usually a made-up sample with the same columns is enough for formula advice. Use that first. If a task requires a real file, follow your organization's data-sharing rules and remove information the task does not need. Work on a copy and confirm which sheet and range the assistant actually analyzed.

How to debug code with AI: a beginner's prompt and test workflow

Reproduce the failure in a small example, include the exact error and state the expected behavior. Ask for a likely cause and the smallest justified change. Run the same failing case after the edit, then test an ordinary case and an edge case. Keep only changes whose behavior you can verify.

In “How to debug code with AI: a beginner's prompt and test workflow”

What should I include when asking ChatGPT to debug code?

Include the language and relevant versions, a small runnable example, the exact input, the error or actual output, and the result you expected. Mention a recent relevant change if you know one. Remove secrets and private records. Ask for one diagnostic check before accepting a broad rewrite.

In “How to debug code with AI: a beginner's prompt and test workflow”

Can I trust code that an AI assistant says it tested?

Look for actual execution output and confirm what environment and cases were used. A predicted result or a written test is not a completed test run. Run the original failing case and relevant regression checks in your own environment before relying on the fix, even if the assistant reports success elsewhere.

In “How to debug code with AI: a beginner's prompt and test workflow”

Should I paste my whole project into an AI chat?

Start with the smallest example that reproduces the failure and the relevant versions or configuration. Too much unrelated material can make the problem harder to isolate, and it may expose information unnecessarily. Expand the context only when a specific missing function, setting or dependency matters to the diagnosis.

In “How to debug code with AI: a beginner's prompt and test workflow”

Why does fixing a coding error with AI sometimes create a new error?

The proposed change may rely on a wrong assumption, a different library version or an incomplete description of expected behavior. Keep a working copy, compare the changes and run tests after each meaningful edit. If the problem moves around, return to a minimal example and gather evidence before trying another rewrite.