Run #769

done rubric v1 session 79393183 gemini-3.7-flash 2026-09-21 02:21

What the student sent the question as it arrived, then each reading of it

OCR of the image source not recorded · did not complete

Empty.

The audit's reading of the question, which is what the Product AI judged against:

None

Audit's summary: The student asked about restriction enzymes. The tutor explained the two main types: Exonuclease and Endonuclease. The tutor clarified that Exonuclease acts on the ends of DNA, while Endonuclease acts within the DNA molecule. The tutor used a mix of Hindi and English in the explanation.

Transcript 240 characters · 300 stripped

Read it no speaker labels — tutor and student run together

Hello, Gulb sir. Good morning, Gulb sir. आपको प्रतिबंधन इंजान देखना है, right? आपको प्रतिबंधन इंजान देखना है, right? बता दे रहा है. देखिए, just take a second. Exonuclease is the end of the nucleus. Thank you for watching. Thank you so much.

What the student sent the question as it arrived, then each reading of it

OCR of the image source not recorded · did not complete

Empty.

The audit's reading of the question, which is what the Product AI judged against:

None

Transcript 240 characters · 300 stripped

Read it no speaker labels — tutor and student run together

Review team

no_issue

Product AI

nothing found

teaching Yes · mistakes No · satisfactory Yes

Each verdict takes your own. Yours is stored against the stack's, so “where is it wrong most often” is a question with an answer.

Work is the right way up The camera view is upside down (180 degrees rotated)

Across frames 3 to 8, the writing on the notepad is completely upside down (180 degrees) from the student's perspective.

seen · 98% confident
Your view
Writing is legible at video size Handwriting is neat and legible

Text defining restriction enzymes ('प्रतिबंधन एंजाइम') and their two types in frames 4-8 is clear and readable despite the inverted orientation.

seen · 90% confident
Your view
The work fills the frame Camera shifts and frames 1 and 9 are severely cut off

Frames 1 and 9 show only a narrow sliver of the page against a dark background, while frames 3-8 capture the page along with background table surface.

seen · 85% confident
Your view
Nothing distracting in shot Workspace is clean with minor background items

Frames 3-8 show a floral desk surface and a box edge, but nothing disruptive to the session.

seen · 90% confident
Your view
The student's problem is established The topic is stated and written at the start

The topic 'प्रतिबंधन एंजाइम' (Restriction Enzymes) is visible in frame 1 and written at the top of the page in frame 3.

seen · 95% confident
Your view
Work proceeds in followable steps Notes build progressively line by line

Frames 3 to 8 show notes developing logically from definition to function to classification.

seen · 95% confident
Your view
It reaches an answer The topic explanation is brought to completion

Frame 8 shows the finished notes covering the definition and two enzyme types before the tutor concludes.

seen · 90% confident
Your view
Something was actually written Not applicable to a camera session

measured
The concept is named before the working The tutor explicitly names the topic at the beginning

The tutor states: 'आपको प्रतिबंधन इंजान देखना है, right?' referring to restriction enzymes.

heard · 95% confident
Your view
Explained rather than jumped to the answer The tutor introduces and explains the concept immediately

The tutor starts by explaining the definition of restriction enzymes as molecular scissors in frame 3.

heard · 85% confident
Your view
Solved step by step, not in one leap Explanation follows a clear step-by-step progression

Working progresses sequentially from what restriction enzymes are to how they cut DNA and their two main types in frames 3-8.

heard · 90% confident
Your view
The session opens properly The tutor greets the student and confirms the topic

The tutor starts with 'Hello, Gulb sir. Good morning, Gulb sir. आपको प्रतिबंधन इंजान देखना है, right?'.

heard · 95% confident
Your view
No long dead air 1 gaps over 6.0s, 6s total (2% of the session), longest 6s

Silence is below -35.0dB — a crude measure on a phone microphone, so treat a low count as weak evidence rather than proof the tutor was talking.

measured
Your view

Observations judged, but not counted as the lab objecting

The camera holds still picture moves 31.2 on average between keyframes, 0.7 hard jumps/min

Consistent with the camera being handheld — a stand would remove this entirely. A handheld phone measured 35.1. No steady-camera session has been measured yet, so the lower band is a guess.

Not counted: Measured over 183 known-bad and 29 known-good camera sessions (2026-09-21): raised on 87% of the bad and 86% of the good, and the motion distributions are the same shape — median frame difference 25 on the bad set, 27 on the good. No threshold on this number separates them, so it is reported and not counted.

measured
Your view
Exposure is usable average brightness 107.8, 17.2% of frame very dark, 0.2% blown out

Blown-out share is reported but not judged: the subject is white paper, so a good session measures 11-24% there too.

Not counted: Same two sets: it raised 6 of 183 known-bad and 2 of 29 known-good sessions. Both too few to mean anything, and the brightness distributions overlap almost exactly. The earlier reading that it 'fires more on clean sessions' was two sessions out of twenty-nine — noise, not an inverted rule.

measured
Your view
Understanding of the concept is checked No conceptual understanding check was conducted

The transcript shows only a confirmation of the topic ('right?') without any conceptual questions asked to the student.

Not counted: Raised on 87% of known-bad and 80% of known-good sessions. The criterion is honest about why in its own `why`: the transcript has no speaker labels, so a real check and a rhetorical 'ok?' are often indistinguishable, and the bar ends up one almost no session clears. Still judged and shown, because reading it beside a recording is how it gets better.

heard
Your view

What the model returned verbatim

{"findings": [
  {"key": "orientation", "verdict": "bad", "headline": "The camera view is upside down (180 degrees rotated)", "evidence": "Across frames 3 to 8, the writing on the notepad is completely upside down (180 degrees) from the student's perspective.", "confidence": 0.98},
  {"key": "legibility", "verdict": "ok", "headline": "Handwriting is neat and legible", "evidence": "Text defining restriction enzymes ('प्रतिबंधन एंजाइम') and their two types in frames 4-8 is clear and readable despite the inverted orientation.", "confidence": 0.9},
  {"key": "framing", "verdict": "weak", "headline": "Camera shifts and frames 1 and 9 are severely cut off", "evidence": "Frames 1 and 9 show only a narrow sliver of the page against a dark background, while frames 3-8 capture the page along with background table surface.", "confidence": 0.85},
  {"key": "workspace", "verdict": "ok", "headline": "Workspace is clean with minor background items", "evidence": "Frames 3-8 show a floral desk surface and a box edge, but nothing disruptive to the session.", "confidence": 0.9},
  {"key": "question_captured", "verdict": "ok", "headline": "The topic is stated and written at the start", "evidence": "The topic 'प्रतिबंधन एंजाइम' (Restriction Enzymes) is visible in frame 1 and written at the top of the page in frame 3.", "confidence": 0.95},
  {"key": "structure", "verdict": "ok", "headline": "Notes build progressively line by line", "evidence": "Frames 3 to 8 show notes developing logically from definition to function to classification.", "confidence": 0.95},
  {"key": "completion", "verdict": "ok", "headline": "The topic explanation is brought to completion", "evidence": "Frame 8 shows the finished notes covering the definition and two enzyme types before the tutor concludes.", "confidence": 0.9},
  {"key": "concept_named", "verdict": "ok", "headline": "The tutor explicitly names the topic at the beginning", "evidence": "The tutor states: 'आपको प्रतिबंधन इंजान देखना है, right?' referring to restriction enzymes.", "confidence": 0.95},
  {"key": "explained_before_solving", "verdict": "ok", "headline": "The tutor introduces and explains the concept immediately", "evidence": "The tutor starts by explaining the definition of restriction enzymes as molecular scissors in frame 3.", "confidence": 0.85},
  {"key": "concept_checked", "verdict": "bad", "headline": "No conceptual understanding check was conducted", "evidence": "The transcript shows only a confirmation of the topic ('right?') without any conceptual questions asked to the student.", "confidence": 0.95},
  {"key": "step_by_step", "verdict": "ok", "headline": "Explanation follows a clear step-by-step progression", "evidence": "Working progresses sequentially from what restriction enzymes are to how they cut DNA and their two main types in frames 3-8.", "confidence": 0.9},
  {"key": "opening", "verdict": "ok", "headline": "The tutor greets the student and confirms the topic", "evidence": "The tutor starts with 'Hello, Gulb sir. Good morning, Gulb sir. आपको प्रतिबंधन इंजान देखना है, right?'.", "confidence": 0.95}
]}

Duration	251s (4.2 min)
Video	1280×800 @ 10.0fps
Audio track	yes
Shot changes
Dead air	1 gaps ≥6.0s · 6.0s total · longest 6.0s
Luma	avg 107.8 · min · max
Medium	camera 30.7% flat background
Content written	7.04% → 2.92% (+74.7)
Transcript	240 characters of speech
Frames	9

Run #769

The recording

What the student sent the question as it arrived, then each reading of it

Transcript 240 characters · 300 stripped

What the student sent the question as it arrived, then each reading of it

Transcript 240 characters · 300 stripped

Review team

Product AI

Measured

What the model returned verbatim

Frames in order, one every ~28s