All Articles
What language should your system instruction be in? Measuring three approaches when most prompts arrive in the user's language
Keep the system instruction in English, or translate it into the user's language? I measured input tokens per language with countTokens, then lined up output-language match and schema compliance to find where nine tokens is enough.
I stopped storing every generation log — three retention tiers and a prompt fingerprint that keeps traceability
I was storing every Gemini API request and response body for debugging. Here is how I moved to three retention tiers plus a prompt fingerprint, and kept the ability to diagnose issues without keeping the text.
I Asked Gemini to Grade My App Store Screenshots. Everything Scored 78–85.
Ask Gemini Vision to grade App Store screenshots out of 100 and good candidates and deliberately broken ones both land at 78–85. Here is how I measured the judge's discrimination power, dropped absolute scoring, and rebuilt it as debiased pairwise comparison.
I Put =GEMINI() in 420 Cells. The Calls Kept Coming on Days I Never Opened the Sheet
A custom function calling Gemini from Sheets looks best the moment it starts working, and it does not grow from there. Here is what the 30-second ceiling, uncontrolled recalculation, and read-only cells actually cost, measured, plus the code to move to a write-back design.
The Table Was There, but the Rows and Columns Weren't — Preserving Docs Structure Before It Reaches Gemini
getBody().getText() flattens Google Docs tables into a column of loose values. Here is what that cost me on a 42-row ledger, the Apps Script extraction layer that keeps the structure, and the acceptance test that keeps it honest.
Sample what you already accepted — an audit budget that catches silent quality drift
A confidence gate only ever looks at output the model hesitated on. Silent drift sinks into the batch that sailed through. Working from a fixed 30-minutes-a-day review budget, this walks through deriving detection time from the binomial, reallocating the same budget across risk strata, and catching slow decay with a cumulative monitor.
A near-miss label won't fix itself on retry — a normalization layer for closed-vocabulary classification
When responseSchema enum returns an out-of-set label, retrying tends to return the same near-miss. From a wallpaper app's 30-category batch, here is the distribution of how labels miss, plus a normalization layer built on an alias table and gemini-embedding-2 nearest-neighbor, with measured results.
Before One Runaway Experiment Drains the Shared Budget: Using AI Studio Spend Caps as Isolation Walls
When you run several Gemini experiments under one billing account, a single runaway loop takes everything else down with it. Here is how I use AI Studio's per-project spend caps as isolation walls, plus a client-side soft ceiling and monthly reconciliation, with working code.
When a Batch Job Sat in RUNNING for Half a Day: Field Notes on Catching Stalls Early with Per-State Dwell Budgets and Record Reconciliation
When a Gemini Batch job stalls quietly under the shadow of the 24-hour SLA, per-state dwell-time budgets and submitted-vs-completed record reconciliation let you name the stall early. Field notes with real operational numbers.
Making TTS seams inaudible: prosody continuity and silence boundaries for long-form narration
Splitting a long article by sentence and synthesizing each piece leaves audible seams: prosody resets at every sentence start, clicks appear at PCM joints, and pauses feel uneven. Four layers -- context priming, zero-crossing edge fades, punctuation-aware silence, and an idempotent seam ledger -- make the joins disappear, with full code.
When Gemini's executed result and its prose disagree on a number — a gate that trusts only code_execution_result
Gemini Code Execution returns the value it actually computed and the sentence describing it as separate parts. Trust the prose and you can inherit a hallucinated number. Here is a verification gate, in working code, that extracts the executed result as the single source of truth and rejects prose that disagrees.
When an Optional Field Comes Back Three Ways: Null, Empty String, and Missing Key in Gemini Structured Output
Optional fields in Gemini structured output drift between null, empty string, and a missing key, and downstream code breaks in three different ways. Here is how I collapse all three into one shape using nullable in responseSchema and a post-output normalization gate, with numbers from a nightly batch.