{"slug": "i-built-an-ai-that-cuts-your-podcast-into-shorts-but-i-didnt-want-it-to-edit", "title": "I Built an AI That Cuts Your Podcast Into Shorts. But I Didn’t Want It to Edit Your Content.", "summary": "An engineer built AI Clip Cutter, an AI tool that turns long-form podcasts into short vertical clips by evaluating candidate moments for hook strength, information density, self-containedness, and emotional tone. The tool explains its choices and keeps trim points editable, ensuring the creator remains in control of the final edit.", "body_md": "The story behind AI Clip Cutter — and why we’re building AI editing around one simple idea: the creator should stay in control.\n\nPress enter or click to view image in full size\n\nThere is an uncomfortable truth about short-form content:\n\nMost creators don’t have a content problem. They have a time problem.\n\nYou can spend an hour recording a podcast.\n\nTwo hours researching.\n\nThree hours having a conversation worth sharing.\n\nAnd then discover that turning that one long video into five genuinely good Shorts is going to take another afternoon.\n\nFinding the moments.\n\nCutting them.\n\nReframing them.\n\nWriting captions.\n\nMaking sure the captions don’t start halfway through a sentence.\n\nChecking whether the clip actually makes sense without the 30 seconds of conversation before it.\n\nThen doing it again.\n\nAnd again.\n\nAnd again.\n\nThat was the problem that led us to build AI Clip Cutter.\n\nAI Clip Cutter\n\nBut there was another question behind it:\n\nWhat if AI didn’t need to replace the editor?\n\nWhat if it could simply do the boring part incredibly well?\n\nThe idea was simple\n\nTake a long-form video.\n\nFind the moments worth sharing.\n\nTurn them into short vertical clips.\n\nAdd captions.\n\nLet the creator decide what gets published.\n\nSounds obvious.\n\nBut once we started building it, we realized that “find the best clips” is not actually a simple problem.\n\nA 60-minute podcast can contain dozens of technically valid 30-second sections.\n\nBut most of them aren’t good Shorts.\n\nSome start in the middle of an argument.\n\nSome need 45 seconds of context.\n\nSome contain interesting information but have no hook.\n\nSome are emotional but say nothing.\n\nAnd some sound incredible when you’re sitting inside the full conversation — but completely confusing when they’re watched alone.\n\nSo we needed the AI to understand something more important than:\n\n“What was said?”\n\nIt needed to understand:\n\n“Would someone want to watch this?”\n\nWe don’t ask AI to pick “interesting” moments\n\nThis was one of our biggest product decisions.\n\nInstead of asking the model to vaguely find “interesting clips,” AI Clip Cutter evaluates candidate moments across four dimensions:\n\nHook strength.\n\nDoes the clip make you want to keep watching?\n\nInformation density.\n\nDoes something meaningful actually happen in those seconds?\n\nSelf-containedness.\n\nCan someone understand the clip without watching the entire podcast?\n\nEmotional tone.\n\nDoes the moment contain surprise, tension, humor, conviction, curiosity, or another strong emotional signal?\n\nThe system generates candidate windows between roughly 18 and 75 seconds, respecting sentence boundaries and natural pauses. It then ranks them and selects a non-overlapping set of clips.\n\nBut here’s the part I’m particularly proud of.\n\nWe don’t just show you:\n\nClip #1–9.2/10\n\nWe show you why the AI picked it.\n\nFor example:\n\n“She reframes the whole argument in one sentence here — a self-contained hook that needs no setup.”\n\nThat’s important.\n\nBecause AI shouldn’t just make decisions.\n\nIt should be able to explain them.\n\nAI gets the first draft. You get the final word.\n\nThere is a dangerous direction AI editing products can take.\n\nUpload your video.\n\nClick a button.\n\nThe AI decides everything.\n\nThe crop.\n\nThe cuts.\n\nThe captions.\n\nAnd eventually, you’re not really editing anymore.\n\nYou’re approving whatever the algorithm decided.\n\nWe wanted something different.\n\nEvery selected clip in AI Clip Cutter comes with its scores and reasoning.\n\nAnd the trim points are editable.\n\nIf the AI picked 00:17:32–00:18:08 and you think the real clip starts two seconds earlier?\n\nChange it.\n\nThe captions are rebuilt around the new timing.\n\nYou remain the editor.\n\nThe AI is your ridiculously fast assistant.\n\nWe also made a slightly controversial decision\n\nWe don’t add b-roll.\n\nWe don’t add random zooms.\n\nWe don’t add transitions.\n\nWe don’t try to make your podcast look like a MrBeast video.\n\nAnd we don’t pretend that every creator wants the same editing style.\n\nThe output is deliberately simple:\n\nYour content.\n\nA clean cut.\n\nA vertical frame.\n\nCaptions.\n\nThat’s it.\n\nBecause sometimes the most powerful thing you can do to a good piece of content is not get in its way.\n\nCaptions are more important than people think\n\nThere is a difference between having captions and having good captions.\n\nA sentence-level subtitle appearing every few seconds is technically a caption.\n\nBut short-form video moves much faster than that.\n\nSo we built the caption system around word-level timestamps.\n\nThe captions know when each word is spoken.\n\nThat allows us to highlight the active word, synchronize the visual rhythm with speech, and create different caption styles without guessing where the words should appear.\n\nThere are currently 18 caption presets, ranging from minimal and cinematic styles to more energetic formats with active-word highlighting.\n\nAnd there’s a small technical detail here that matters:\n\nWhen you change a clip’s trim points, we don’t simply slap the old captions onto the new video.\n\nWe regenerate the caption groups against the new timeline.\n\nIt sounds like a tiny feature.\n\nIt isn’t.\n\nIt’s the kind of detail that determines whether an automated editing product feels genuinely useful or feels like a demo.\n\nWe built for creators in India first\n\nThere’s another part of this product that I’m especially excited about.\n\nAI creator tools are becoming global.\n\nBut the pricing infrastructure hasn’t always caught up.\n\nA lot of creator software is priced in dollars, even when the creator is earning and spending in rupees.\n\nThat creates unnecessary friction.\n\nSo AI Clip Cutter is priced in INR, with payments through Razorpay.\n\nOur current paid plans start at ₹399, with higher tiers at ₹899 and ₹1,999. Credits don’t expire, and the free signup includes 10 credits with no card required.\n\nThis isn’t just a pricing decision.\n\nSubscribe to the Medium newsletter\n\nIt’s a product philosophy.\n\nIf you’re a creator who publishes three videos this month and none next month, your unused credits shouldn’t disappear just because a calendar flipped.\n\nThe economics are intentionally simple\n\nWe don’t want you calculating complicated upload quotas.\n\nThere are basically two things you pay for:\n\nAnalysis: 1 credit per minute of source video.\n\nExport: 1 credit per finished clip.\n\nThat’s it.\n\nSo if you analyze a 30-minute podcast and export four clips:\n\n30 credits for analysis.\n\n4 credits for exports.\n\n34 credits total.\n\nThe model is transparent before you press the button.\n\nNo surprise invoice.\n\nNo mysterious “AI processing units.”\n\nNo pretending that a 30-minute video and a 2-minute video cost the same to process.\n\nAnd yes, there are things our AI can’t do\n\nThis is probably the most important thing I can say about building AI products:\n\nDon’t hide the limitations.\n\nAI Clip Cutter currently doesn’t have face tracking.\n\nIt doesn’t automatically add b-roll.\n\nIt doesn’t add transitions or zooms.\n\nIt doesn’t have speaker diarisation.\n\nVertical reframing currently tops out at 1080×1920.\n\nAnd although captions can handle many languages, our clip scoring and hook-writing are currently strongest for English-language content.\n\nWe’re not embarrassed by that.\n\nIn fact, I’d rather tell you exactly what the product doesn’t do than promise an imaginary future version of it.\n\nBecause the goal isn’t to build the longest feature list.\n\nThe goal is to build something people actually use.\n\nThe bigger opportunity isn’t “AI video editing”\n\nI think that phrase is too broad.\n\nThe interesting opportunity is something much more specific:\n\nTurning existing knowledge into more opportunities to be discovered.\n\nThink about how much valuable content already exists.\n\nPodcasts.\n\nInterviews.\n\nFounder conversations.\n\nCourses.\n\nWebinars.\n\nLivestreams.\n\nCustomer calls.\n\nConference talks.\n\nEducational videos.\n\nThere are millions of hours of long-form content being created.\n\nMost of it will never be turned into short-form content.\n\nNot because it isn’t valuable.\n\nBecause somebody has to sit there and find the moments.\n\nThat’s the bottleneck.\n\nAnd that is exactly where AI is useful.\n\nOne podcast can become an entire content engine\n\nImagine recording one 60-minute conversation.\n\nTraditionally, that becomes:\n\n1 long video.\n\nWith a good clipping workflow, it can become:\n\n1 YouTube episode\n\nplus\n\n5–10 short videos\n\nplus\n\nmultiple hooks\n\nplus\n\ncaptioned social posts\n\nplus\n\nideas for future episodes\n\nThe original conversation doesn’t change.\n\nYou’re simply extracting more value from something you already created.\n\nThat’s the part of AI that excites me.\n\nNot replacing creativity.\n\nCompounding it.\n\nWe are not trying to make creators obsolete\n\nThis distinction matters.\n\nI don’t believe the future of creative work is:\n\nHumans create → AI replaces them.\n\nI think it’s closer to:\n\nHumans create → AI removes the repetitive work → humans make better decisions.\n\nThe creator should decide what they stand for.\n\nThe creator should decide what feels authentic.\n\nThe creator should decide what gets published.\n\nThe AI should handle the parts that humans shouldn’t have to spend hours doing.\n\nFinding timestamps is not creativity.\n\nGenerating 30 caption variations is not creativity.\n\nReframing a video from 16:9 to 9:16 is not creativity.\n\nSearching through a one-hour podcast for the strongest 40-second moments is useful work — but it’s also incredibly repetitive.\n\nThat’s where we want AI.\n\nWe’re building the boring parts of content creation\n\nThat might sound like a strange startup pitch.\n\nBut I think it’s a powerful one.\n\nThe future of creator software isn’t necessarily about adding more effects.", "url": "https://wpnews.pro/news/i-built-an-ai-that-cuts-your-podcast-into-shorts-but-i-didnt-want-it-to-edit", "canonical_source": "https://dev.to/munjapara/i-built-an-ai-that-cuts-your-podcast-into-shorts-but-i-didnt-want-it-to-edit-your-content-56mg", "published_at": "2026-08-16 03:50:56+00:00", "updated_at": "2026-08-16 04:10:56.528565+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-products", "ai-tools", "generative-ai", "natural-language-processing"], "entities": ["AI Clip Cutter"], "alternates": {"html": "https://wpnews.pro/news/i-built-an-ai-that-cuts-your-podcast-into-shorts-but-i-didnt-want-it-to-edit", "markdown": "https://wpnews.pro/news/i-built-an-ai-that-cuts-your-podcast-into-shorts-but-i-didnt-want-it-to-edit.md", "text": "https://wpnews.pro/news/i-built-an-ai-that-cuts-your-podcast-into-shorts-but-i-didnt-want-it-to-edit.txt", "jsonld": "https://wpnews.pro/news/i-built-an-ai-that-cuts-your-podcast-into-shorts-but-i-didnt-want-it-to-edit.jsonld"}}