Turning vague client notes into changes
A translation table from client language to testable production changes, plus the three questions that resolve a vague note before you spend attempts.
9 min read
Building a UGC rate card that prices usage rights
Usage rights, not production, are where UGC margin lives. A rate card separating base fee from 30-day, 90-day and unlimited buyout, plus enforcing clauses.
9 min read
The upscale broke text that was already correct
Generative upscalers re-draw letterforms and will respell a word that was already right. Mask text out of the pass, or drop denoise on text-bearing images.
9 min read
Variable frame rate breaks caption timing
Phone footage and screen recordings often carry variable frame rate. Detect VFR, transcode to a constant rate, then caption so cues stop drifting.
9 min read
9:16 keeps cutting your subject off at the waist
Aspect ratio is a composition input, not a crop. Tall frames need shot-type language that reaches the feet, plus recovery options short of a re-roll.
9 min read
The outfit changes between shot 2 and shot 3
Costume drift is identity drift you only catch at assembly. Reuse one wardrobe block verbatim, and know which garment details need a still instead of text.
9 min read
What an AI-use audit clause covers
Vague audit rights over AI use turn into open-ended discovery. A scoped clause, plus the record set that makes the whole audit take an hour.
9 min read
What goes in a $1,500 local video retainer
A defensible entry tier: deliverable count, format mix, revision cap, reporting cadence, four written exclusions, and the triggers for the next tier.
9 min read
What belongs in a prompt pack people don't refund
A list of prompts is not a product. The contents template that separates a pack buyers keep from one they refund within the hour.
9 min read
What Test & Compare won't let you test
Shorts, Made for Kids, private, and age-restricted videos cannot run Test & Compare. The eligibility matrix, what stops a live test, and workarounds.
9 min read
What the ELVIS Act protects in a voice
Tennessee made voice a protected personal right and reached the tool layer, not only publishers. Why building with voice clones carries statutory exposure.
7 min read
What to write instead of a negative prompt
Negative prompts are a no-op on flow-matching models and leak the negated noun elsewhere. Positive re-specification and the three constraints that replaced it.
9 min read
What 'you own the output' actually means
Vendor terms can assign you an output file. They cannot make it copyrightable. How to separate contractual ownership from protectability before you ship.
9 min read
When a 4K export is worth the credits
Most social delivery wastes the 4K premium. The three cases that earn it, the reframe arithmetic behind them, and a per-destination resolution rule.
8 min read
When the agent picks the wrong sub-agent
Misrouting is usually ambiguous phrasing, not a broken agent. Word the first sentence so only one of the six domains can claim the job.
9 min read
Who indemnifies what in an AI contract
Indemnity splits two ways: you cover tool terms-of-service breaches, the client covers distribution and labelling. A two-way split to drop into an agreement.
9 min read
Who owns the prompt library you built?
Prompts, reference sets and workflow graphs built on a client job are unassigned in most contracts. Clause language that keeps method and gives outputs.
9 min read
A batch of eight is not eight single runs
Batch size changes the noise each sample gets. A batched call and eight parallel calls diverge, even with the same seeds. Know when that variance is harmless.
9 min read
Why C2PA manifests die on upload
A C2PA manifest is a hash over exact bytes, so re-encodes and screenshots kill it by design. How to find the hop in your pipeline where provenance is lost.
8 min read
Why the editor rejects local file paths
Clip sources must be reachable over HTTPS, so file:// paths fail immediately. Upload or attach first, then render from the hosted URL.
7 min read
Why your clip shimmers, and how to stop it
Shimmer is usually a frame-rate problem: too many frames starve temporal context. Rates that hold, where the deflicker pass goes, and how to spot compression.
9 min read
How to fill a 4,500-token prompt window
Long-window image models take far more instruction than most prompts supply. A block structure for long prompts, and where extra tokens stop mattering.
8 min read
Writing acceptance criteria clients can't argue
Turn 'approved' from a feeling into a test: technical spec, binary brand checks, a bounded subjective layer, a named approver, and a deemed-accepted window.
9 min read
Your cloned voice sounds nothing like you
Clone quality is set by the reference recording, not the model. Source-audio requirements and the three defects that guarantee a bad clone.
9 min read