~60s How it works

No black box.
Every roast, explained.

We don't believe in mystery scoring. Here's exactly what happens between upload and insight.

01

You upload your video

Pick any short-form video from your camera roll — up to 3 minutes, 50MB. The app streams it to our servers in chunks so the upload is fast even on LTE.

Format
MP4 / MOV, vertical or horizontal
Duration cap
3 minutes
Size cap
50MB
Upload time
~5-20 seconds on modern connections
02

We extract and transcribe

ffmpeg extracts 10 keyframes at even intervals across the video. OpenAI Whisper transcribes the full audio track with a confidence filter that rejects hallucinations (music-only clips produce zero transcript, not fake dialogue).

Keyframes
10 images, 720p downsampled
Transcription
OpenAI Whisper, segment confidence filtered
Hallucination defense
3-layer blacklist + silence check + confidence gate
Time
~8-15 seconds
03

Claude analyzes

Anthropic's Claude Sonnet 4 receives the keyframes and transcript. It scores 7 categories with defined rubrics, identifies 8-12 specific passes/fails, writes 3 savage roast lines referencing exact moments, and — if you provided performance data — reconciles the score against reality.

Model
Claude Sonnet 4 (vision + text)
Categories scored
Hook · Value · Pacing · CTA · Visual · Trend · Audio
Output
Scores, findings, roast lines, visual reading, per-line scoring
Time
~25-45 seconds
04

You get the grade

An A-to-F letter grade, overall score out of 100, and a MID / UNMID verdict. Tabbed breakdown: the roast itself, category scores, detailed findings, your ideals comparison (if seeded), and the proof — the transcript, keyframes, and line scores that back every claim.

Grade reveal
Anticipation-designed with dopamine trigger
Tabs
Roast · Scores · Findings · vs Ideals · Proof
Peak-end card
"Your weakest category is X — here's how to master it"
Share
9:16 image card for Stories / Reels / TikTok
05

You fix, re-upload, rescore

Apply the notes, shoot a v2, upload again. Your score climbs (or doesn't). Over time, you build the habit of unmidifying every video before you post it. That's the real product.

Streak tracking
Daily roasting = visible streak = loss aversion
Cheat Sheet
Evergreen fundamentals when you don't have a video to roast
Ideal Posts
Your curated benchmark library for personalized comparison
History
Scores trend over time — prove you're getting less mid

Radical transparency

Here's how to fake our scoring with ChatGPT.
We dare you.

We publish our scoring rubric openly. Competitors can't do this without killing their product. We can because the value isn't the formula — it's the execution, the fix loop, and the library.

Read the rubric

Ready

Stop posting blind.