There are editors in the Bollywood industry who get paid crores of rupees and have never once clicked into the actual editing software.1 Assistants make the first cut. Assistants make the second cut. The editor with the enormous fee walks in afterward, watches, and adjusts. That's the whole job, from the outside. From the outside it looks like almost nothing — someone watching footage other people already cut, occasionally saying "move this" or "cut that."
So what, exactly, is anyone paying crores for?
Krisang describes what he actually tells his own editor, and it's not a technical brief at all: "Just feel your edits. That's all."2 No shot-list logic, no pacing formula, no rule about cut-on-action or match-on-motion. One instruction, and it's an instruction to feel something, not to do something. His conclusion follows directly: "Editing is not a technical skill. It is an emotional skill. If you cannot feel your edit, you cannot be a good editor."2
This reframes what the crore-paid Bollywood editor is actually selling. Not the cutting — assistants do the cutting, and cutting is learnable in months. What the top editor sells is the capacity to feel the shot, to feel the scene on the screen, and to know from that feeling — not from a rule, not from a formula — whether it's working.3 "They are capable of feeling the shot. They are capable of feeling the scene on the screen. They're getting paid to feel stuff."3 The technical execution is delegable. The felt judgment about whether the execution is right is not, and that's the part with the price tag on it.
The obvious objection: isn't "it just feels right" a way of avoiding the harder work of explaining your reasoning? Krisang's answer runs the other direction. Feeling isn't instead of precision — it's what precision is built on top of. "If I don't like a design I should be able to tell you precisely what is it about that design I'm not liking. That is the skill you should be developing."4 The sequence matters: feel first, then translate the feeling into words precise enough to act on. Skip the feeling step and go straight to analysis, and you get technically defensible judgments that miss whatever the audience would have actually responded to. Skip the translation step and stay purely in feeling, and you can't direct anyone else to fix what's wrong. Both steps are required; only the first one is the skill nobody teaches.
You're reviewing a rough cut. The old habit is to check it against a rulebook in your head — is the pacing right, does this scene run long, does the cut motivate the next one. Try something different: watch it and notice, physically, whether anything in you tightens, flattens, or loses interest. Don't reach for the reason yet. Just clock the sensation. Then go back and ask what specifically produced it — which cut, which beat, which held-too-long shot. The reason comes second. If you go looking for the reason first, you'll find a plausible one whether or not it's the real one.
You're the one being handed the cut, not the one making it. An editor shows you three options for the opening ten seconds. The old habit is to evaluate them against a checklist — hook, pacing, clarity. Instead, watch all three back to back and notice which one you feel the pull of before you've consciously decided anything. Then interrogate that pull until you can hand back a note more useful than "I liked the second one better" — a note that names exactly what the second one is doing that the others aren't.
You're training someone else to edit for you. Don't hand them a style guide first. Sit with them while they cut, and when something's off, don't reach for the rule — reach for the sensation, and describe the sensation out loud, in front of them, until they start noticing the same thing without you pointing at it. That's the actual transmission. The rules, if you ever write them down, come after the sensation has already been trained into both of you.
Krisang is blunt about why this stays invisible: "All you have been taught is to distrust your feeling. And people are getting paid in lakhs and crores for that."5 The claim is specific and a little uncomfortable — not that feeling is undervalued in some vague cultural sense, but that there is a live, functioning market paying premium rates for exactly the capacity most training actively suppresses. The Bollywood editor's fee isn't evidence that feeling should matter in creative judgment. It's evidence that it already does, whether or not anyone's curriculum admits it.
Krisang extends the same instrument to AI-generated output without treating it as a different problem: "With AI now your skill is basically having a taste. You should be able to discern what is good and bad. Good and bad is not an intellectual exercise, it's a feeling."6 Reading an AI draft, the question isn't "is this logically correct" — it usually is. The question is "am I feeling anything," and if the answer is no, or the wrong thing, that's the actual signal something needs to change, prior to any argument about why.
How do you know you're looking at good taste rather than a good rule followed correctly? The tell, on Krisang's account, isn't in the output — it's in the response time. A felt judgment arrives before you can articulate it; you know something's off in the half-second before you can say what. A rule-based judgment arrives after you've run the checklist. Neither is fake, but only one of them is what the top editors are being paid for, because the checklist can be handed to an assistant. The half-second can't.
The Bollywood-editor economics claim is asserted, not sourced with names or figures — it's Krisang's account of an industry practice, offered as evidence for the feeling-not-technique thesis. [PARAPHRASED] It's plausible (division of technical and evaluative labor is common in high-budget production) but unverified as stated.
Tension worth naming: "just feel it" is genuinely difficult to teach or transmit as instruction — it's easy to state and hard to operationalize, which is exactly the gap the Implementation Workflow above tries to bridge with a concrete sequence (sensation first, translation second). Whether that sequence actually produces trainable taste, or whether taste is closer to an aptitude some people have and others don't, is left open by the source.
Put Krisang's editing claim next to Bharata's sattva doctrine and something interesting happens: two traditions, two thousand years apart, independently landing on the same ranking principle — that the highest-tier practitioner is defined by an internal, involuntary quality rather than external technical correctness. Bharata ranks actors by whether their bodily signs of emotion arise from actually being in the state (sattvātirikta, superior) versus being technically summoned from outside (asattva, inferior) — and insists this quality is invisible, nameable only by its effects.7 Krisang ranks editors the same way: technically correct cuts that don't come from felt judgment are worth an assistant's wage; cuts that do come from felt judgment are worth crores. Neither tradition can fully specify what the internal quality is. Both can specify, with confidence, what it produces and what its absence looks like. The convergence suggests the gap isn't a failure of either tradition to think hard enough — it may be a structural feature of any domain where the highest-value judgment is bodily before it's verbal.
Business — Taste, Judgment, Labor Framework. Simmonds's framework names Judgment as one of three separable roles in AI-assisted creative work — the "rigorous editorial evaluation of the AI's Labor against the uncompromising standard of the human's Taste" — and gives it a diagnostic: the creator judging well is "completely detached from the ego of having written it," able to discard weak drafts without pain because they didn't sink the hours into producing them. What that framework names but doesn't explain is how Judgment actually operates as a cognitive act. It describes the role, the position in the pipeline, the psychological permission structure — but not the mechanism by which a human tells good AI output from bad. Krisang supplies exactly that missing mechanism: Judgment is felt before it's reasoned, in exactly the same way the Bollywood editor feels a cut before they can justify it. Read together, a fuller picture forms than either page holds alone. The TJL framework tells you Judgment must stay human and names why (ego-detachment, taste-fidelity). Krisang tells you what the human is actually doing in that seat — not applying a checklist, but running a felt-response test and then translating the result into an instruction the Labor layer (bot or assistant) can act on. The TJL framework's own diagnostic-signs section ("the output fundamentally feels like it belongs to the creator") uses the word feels without examining it. Krisang's page is the examination.
Eastern spirituality — Sattva — Temperament and Method Acting. Bharata's three-tier ranking of performance — sattvātirikta (the actor genuinely enters the state, superior), samasattva (partial access, middling), asattva (technique only, inferior) — hands the editing-as-emotional-skill claim a vocabulary it doesn't have on its own. Krisang asserts a two-tier distinction (technical cutting vs. felt judgment) and a market fact (the felt-judgment tier gets paid crores). Bharata's third tier — samasattva, the competent-but-not-transformative middle — names something Krisang's binary misses: most working editors aren't purely technical and aren't purely felt-judgment; they're partially in the state, producing work that's fine but not the thing crores get paid for. The handshake sharpens the diagnostic in both directions. From Bharata's side: the audience "can tell" the difference between sattvātirikta and asattva performance even when every visible sign is technically correct, which is the theatrical version of Krisang's claim that a felt cut and a technically correct cut can look almost identical and land completely differently. From Krisang's side: the Bollywood-editor economy is empirical proof, outside the theater, that Bharata's centuries-old ranking tracks a real and monetizable skill differential — not just a theory about stagecraft, but a description of how markets already price felt judgment when they encounter it.
Sharpest implication. If editing (and, by extension, most creative judgment) is an emotional skill wearing a technical costume, then most craft education is teaching the wrong layer — training people to defend judgments with rules after the fact, while leaving the actual judgment-generating capacity, the felt response, entirely untrained. You can graduate from an editing program having mastered every technical rule and still be worth an assistant's wage, because the rules were never the scarce resource.
Generative questions.