Documenting A Product Launch

I’m the founder of Oru, a hardware startup in Vermont. We make a connected woodstove monitor: a temperature probe on your stove’s flue pipe, wirelessly connected to a small lamp made of walnut and frosted glass that glows different colors depending on what your fire is doing. One color means the fire is cruising. It shifts when the stove wants wood. Another color means it’s running too hot. You don’t check an app. You just see it, from across the room, even when your hands are full. I’m launching a Kickstarter for it in October. Between now and then, I have to manufacture a thousand units, and I’m documenting the whole run in a YouTube series called First Snow. The name comes from the deadline: wood burning season in Vermont starts when the nights drop below fifty degrees, and no Kickstarter deadline matters more than the weather’s. I need an editor for episode 1, which is fully scripted, fully shot, and ready for assembly. If episode 1 goes well, this becomes recurring weekly work through October and likely beyond. The longer version, because it’s the material you’ll be cutting Three years ago I bought a house in the middle of the woods in Vermont. I’m a city kid. The house came with two woodstoves, and I had never used one in my life. My first winter, a tree fell on the house. That’s the level of “didn’t know what I was getting into” we’re working with, and yes, there’s a photo, and yes, it’s in the video. That first winter I had to learn, very quickly, with a pregnant wife at home, how to heat a house the old fashioned way. Nobody teaches you this. There’s a cast-iron box, a fire inside it, and a feeling in your gut that you’re doing it wrong. I checked the stove constantly. Walked over, looked at the fire, walked away, came back ten minutes later. The stove was our primary heat source and it was literally and figuratively a black box. I’m an engineer. I’ve spent my career building hardware and software. So one weekend I did the dumbest reasonable thing available: I stuck a temperature probe on the flue, wired it to a Raspberry Pi, and had it text me what was going on. Suddenly I could see the fire. When it was ramping. When it was cruising. When it needed to be reloaded. The texts worked great, technically. Practically, my wife was chasing a toddler around and did not want to take out her phone. So we turned the device into a lamp. That lamp became a product. I posted it to the woodstove corner of Reddit expecting nothing, and strangers started asking to buy one. Real comments, real strangers: “I’ll take 3. Where do I send money?” Which was a problem, because I knew how to write software and solder a board, but a thing people pay for has to be built. So I learned to woodwork. I learned to run a CNC. I hand-built the first batch at Generator, the makerspace in Burlington where I still work every day, and they sold out. People all across the country bought one. This spring, the product won LaunchVT, a statewide startup competition. Now I’m scaling from hand-built batches to a real production run of a thousand units, with a Kickstarter in October and winter closing in. That’s the series. Episode 1 tells the origin story you just read and starts the countdown clock. Every subsequent episode documents one week of the run: manufacturing, the Kickstarter, the mistakes, all of it. My favorite thing a customer ever said about the product came from a guy I had beers with. He called it “a peaceful reminder to stoke the fire.” That’s the emotional register of this whole series. Warm, calm, honest, quietly funny. No hype, no hustle-culture energy, no fear. Why I’m hiring an editor instead of doing it myself I cut the rough intro myself and it taught me two things. First, I can see what the edit should be. Second, I am slow at making it be that. I’m running a hardware company, a manufacturing ramp, and a crowdfunding launch at the same time. The twenty hours I’d spend fighting a timeline are twenty hours the company needs elsewhere, and a good editor will make the video better than I would anyway. What I bring to the collaboration is unusual, though, and it’s worth being clear about: I am not handing you a folder of clips and a vibe. I’m handing you a complete edit script. Every VO line in the episode is mapped to its shot, its source file, its timecode range, and its treatment. You will never wonder what goes where. Your craft goes into pacing, rhythm, sound, and polish, not into guessing my intent. PART 2: THE SERIES AND ITS AUDIENCE What First Snow is A founder-led documentary series about building a hardware product against a hard seasonal deadline. The format is narrated process video: my voice over real footage, my face appearing rarely and deliberately. Episode 1 is the origin story at 6 to 7 minutes. Episodes after that are weekly, shorter, and built from one master shoot per week plus screen captures and bench footage. Each master episode also yields 3 to 4 vertical shorts for Instagram Reels, TikTok, and YouTube Shorts. The shorts are cut from the same timeline, which is why every shot in the series is framed with the action in the center third: the same take has to survive a 9:16 crop. Who watches this The audience is woodstove owners and the maker/homestead YouTube crowd. Think of the people who watch Blacktail Studio build a table for forty minutes, or Gold Shaw Farm walk a fence line, or StuckAtPrototype turn a breadboard into a product. They are patient, they are allergic to hype, and they have extremely sensitive detectors for anything fake. That last point shapes every rule in this project. This audience will freeze-frame a suspicious shot and put it in the comments. Authenticity isn’t a brand value we aspire to, it’s a production constraint we enforce. Which brings me to the rules. The production rules (non-negotiable, and your protection as much as mine) No AI-generated imagery ever ships. I use AI clips as previz to plan sequences, and you may see them in reference materials. They are timing placeholders. Every frame in the published video is real footage, real photos, or real screen captures. If you’re ever unsure whether an asset is cleared to ship, ask. “Is this real?” is always a welcome question on this project. Chronology is sacred. The story spans three winters, and nothing appears on screen before it exists in the story. The Raspberry Pi era shows text messages, never the app that didn’t exist yet. The 3D-printed lamp appears before the wooden lamp. The current product appears only in present-day scenes. The edit script enforces this; your job is to not break it while cutting, and to flag it if you spot a violation I missed. The fire is real. All flame footage is from my actual stove, shot in January. We never light a fire in July for b-roll, and we never use stock fire. My face stays off-screen until the final section, with one sanctioned exception in the cold open. The reveal at the end is earned by its absence. This is a structural device, not shyness, and the edit needs to protect it. Real comments only. The Reddit sequence uses screenshots of actual comments from actual posts, rendered as graphic cards. Nothing is paraphrased, nothing is invented. Numbers on screen and in VO match the publish date. The countdown says “thirteen weeks” only if it’s actually thirteen weeks when the video goes live. If the edit slips a week, the VO line gets re-recorded, not fudged. If these rules sound restrictive, here’s the reframe: they’re why the project is satisfying to cut. Every asset you touch is genuine, which means every emotional beat you land is earned. You will not be asked to make something hollow look sincere. The material is sincere. Your job is to make it land. PART 3: EPISODE 1 IN DETAIL This section walks the full structure so you know exactly what you’re signing up to cut. Timings are targets, not handcuffs. The episode runs about 6:30. The audio arc (read this first, it shapes everything) The episode’s sound follows a single build: silence, then a spark, then momentum, then full speed, then quiet. Specifically: The cold open has no music at all. Thirteen seconds of real sound: solder sizzle, parts clicking, room tone, my voice. The first note of music, a single fingerpicked guitar, lands exactly on the cut to the title card. Act 1 rides that sparse guitar, barely audible. Act 2 adds a light rhythm, then stomp-and-clap percussion when the Reddit montage hits. Act 3 runs the full driving folk track, the most energy in the video, peaking on the final countdown lines. Then everything strips away for the close: quiet guitar, then near-silence, my face on camera for the first time, talking to a friend. Inside that build there are two hard stops where the music cuts dead for a full second. One lands on the best Reddit comment. One breathes before the lamp reveal. A build that never pauses has no punctuation; these two silences are the punctuation, and they need to be exact. The music is instrumental acoustic Americana. Fingerpicked guitar, banjo, upright bass, stomps and claps for percussion. No vocals, and specifically no gang-vocal “hey!” folk, which reads as an insurance commercial. Warm, woody, a little sparse. Vermont in audio form. I’ll approve final track selections; you’ll propose them from a library we agree on (I have Epidemic Sound or we use the YouTube Audio Library’s folk section). Cold open (0:00 to 0:13, no music) Fast cuts, roughly one per second: macro soldering, the temperature probe in my fingers, the workbench, then the product glowing through its color states with a real fire behind it. VO carries the stakes: wood burning season starts when the nights drop below fifty, that’s thirteen weeks away, I need to build a thousand of these before the first snow. The section ends with a settle: the pace stops, one held shot, and the line “Follow along to see if I make it,” then a breath of silence before the title. The cut rhythm here is word-locked, not beat-locked, since there’s no music. “These” lands on the product. “First snow” lands on winter imagery. The contrast between the fast open and the dead-stop settle is the whole design; if the settle doesn’t feel like an exhale, the open isn’t done. Title beat (0:13 to 0:20) “First Snow” in serif type over full-frame real fire, subtitle “Pt 1. Teaching my woodstove to talk.” The first guitar note lands on this cut. VO: “I’m Neil. I make woodstoves talk.” Short, warm, done. Act 1A: The house (0:20 to 1:00) Photo-driven storytelling with slow push-ins. My real house in spring (“bought it three years ago with no idea what I was getting into”), then the same house in winter with a fallen tree on it. The edit holds the wide long enough for the viewer to find the tree themselves, then punches in on it exactly on the line “I told you I didn’t know what I was getting into.” The punch-in is the punchline, and it’s followed by a full beat of silence. Comedy through cutting, not through jokes. Then the stoves: two static shots cut on the word “two,” a slow push into a cold empty firebox on “I’m a city kid,” and hands stacking firewood on the “learn the old fashioned way” line. Every image is real: my house, my stoves, my firewood. Act 1B: The black box and the hack (1:00 to 2:30) This is the emotional core, and it has a specific visual logic: the fire stays illegible until the exact moment the story says I could finally see it. The act opens on archive footage of my actual stove burning, but the shots are dark, smoky, hard to read. Blue flame wisps, ember glow, glass you have to squint through. That’s deliberate: the viewer experiences the fire the way I did, as a thing you can see but not understand. A close-up of the sad little analog magnetic thermometer on the flue pipe lands on “a feeling in your gut that you’re doing it wrong,” because that gauge was the state of the art. The checking-anxiety beat uses handheld archive footage that drifts up to the stove glass like a nervous person approaching, twice, cut with different angles. Then the darkest, least readable shot under “it was literally and figuratively a black box.” Then a hard cut, gloom to bright workbench: the bare Raspberry Pi in my fingers. The cut itself is the joke. The probe goes on the flue (a one-second shot of the old analog gauge, then the probe clamped in its place: old way, new way). Then the payoff: a real iPhone screen recording scrolls through months of text messages from the Pi. “Your burn has started: 213°F.” “ALERT: Your stove’s temperature is rising really fast.” “Your stove is cooling, it might be time to reload.” On “And suddenly I could see the fire,” the episode shows its first clean, readable, full-speed burn. The footage quality arc IS the story arc. Then three text bubbles punch in, one per phrase: ramping, cruising, reloaded. Act 2A: The turn (2:30 to 3:10) The device worked, technically. Practically, my wife was chasing a toddler and didn’t want to take out her phone. The footage here is gold: real archive of the baby gate corralled around the burning stove (one image that tells the entire story), toddler toys on the floor, and a staged-but-real iPhone notification arriving on a counter, sitting unread, and going dark. Then the thesis cut of the episode: the dark phone screen cuts directly to the first lamp glowing warm on the same kitchen counter. Dark to light, failure to answer. “So we turned that device into a lamp.” A wonder cue swells. The lamp’s color states get taught one phrase at a time, and the act closes on a real winter living-room wide with the lamp glowing across the room. Act 2B: The Reddit moment (3:10 to 3:40) This is the most stylized sequence and the closest to the Scott Yu-Jan reference. A real screen recording of the actual Reddit post scrolls plainly. Then comment cards, pre-rendered graphics of real comments, pop onto a dark background one per music beat, stacking with slight rotations. “This is freaking brilliant!” “DM me I want one of these.” “This is going on my Christmas list.” Then the music stops dead. One card alone on screen: “I’ll take 3. Where do I send money?” A full second of silence. That silence is the punchline of the first three minutes of the video. Music resumes on a filmed shot, real hand holding a real phone, thumb scrolling the actual thread, warm lamplight. VO: “Which was a problem.” Act 2C: It had to be built (3:40 to 4:30) Beat-cut montage: circuit boards on the bench, CNC running, sawdust, the makerspace, batch assembly photos, the finished wooden lamp glowing. “They sold out.” A live map with dots across the country. Then the tempo dips for the customer quote, which fades in as on-screen text over the lamp glowing in the dark: “a peaceful reminder to stoke the fire.” Hold it two seconds past the VO. Let it sit. Bridge (4:30 to 4:45) One shot: every version of the product in a row on the workbench. The bare Pi rig, the 3D-printed lamp, the wooden lamp, the current product. A slow lateral move. “This spring, the fourth version of that weekend hack won a startup competition here in Vermont.” The whole journey in one frame, right before the countdown says how far is left to go. Act 3: The countdown (4:45 to 6:00) The fastest cutting in the video, fully beat-locked to the driving folk track. Rapid inserts: CAD models spinning, quote folders, calendars, the shop today. The Kickstarter date. Then the tempo opens up for the stakes: “The deadline underneath all of it isn’t mine. It’s the weather’s. If I miss October, I don’t miss a deadline. I miss the winter.” Real snow footage breathes at 2 to 3 seconds per shot against the fast section around it. Then the final run: the episode’s fastest sequence, echoing the cold open’s shots, one cut per beat, music peaking, landing on “So: thirteen weeks. A thousand units. One person. One makerspace. One iPhone.” Close (6:00 to 6:45) Everything strips away. Quiet guitar, then almost nothing. My face on camera for the first time in the episode, at the bench, talking like a person and not a pitch: I’m going to document all of it, follow along, and if you want one of the first thousand there’s a reserve list at getoru.com. End card: First Snow, a small glowing cube on black, “See you tomorrow.” One wood-on-wood click, cut to black. PART 4: WHAT YOU GET FROM ME The edit script The core handoff document is an edit script that maps the entire episode line by line. Each row contains: the exact VO line, the shot that goes under it, the source file and timecode range, and the treatment (speed, punch-in, hold, transition, sound note). A sample row so you know the format: VO: “…and a feeling in your gut that you’re doing it wrong.” Shot: Analog flue thermometer with red magnet, close. Source: IMG_6459.MOV @ 0:18–0:21. Treatment: Static hold. The sad gauge is the punchline. There are about fifty rows like that covering the full episode. You will have creative latitude inside each row (exact frame choice, cut timing, transition handling) and the standing invitation to propose a better idea whenever you see one. The script is the floor, not the ceiling. The footage and asset package Everything is organized and labeled before you receive it: Bench and product footage, iPhone 4K, shot specifically for this episode: soldering macros, probe assembly, the product cycling its color states in front of a real fire, workshop coverage. Winter archive footage from January: the stove burning (wides and close-ups), handheld fire-glass shots, the baby gate around the stove, the lamp glowing across a real living room, and a live demo of the probe changing the lamp’s color. Photo archive: the house in spring, the house in winter with the tree on it, the 3D-printed lamp prototype on the kitchen counter, circuit board development shots, batch assembly photos. A finished 21-second screen capture of the Pi-era text message thread scrolling (already status-bar-cleaned and ready to place). Seven pre-rendered Reddit comment cards as transparent PNGs, ordered by escalation, with consistent shadows. Screen recordings: the actual Reddit posts scrolling, the fleet map, CAD. A scratch VO track for the full script, recorded by me, good enough to cut against. The final VO gets recorded after picture lock, matched to your locked timing, and swapped in. Foley/ambience: room tone and working sound recorded at each location for the silent cold open. Resolution note: the master timeline is 4K. Some archive footage is 1080p; the script marks which sources can take punch-ins (4K) and which run full-frame (1080p, passes as archive texture). References You’ll get links to the three reference videos with timestamps and notes on exactly what we’re borrowing from each: the beat-locked montage and hard-stop grammar from Scott Yu-Jan, the calm narrated pacing and quiet confidence from Blacktail Studio, and the seasonal-clock storytelling from Gold Shaw Farm. You’ll also get my previz cuts of two sequences, which show intended timing even though their imagery never ships. PART 5: THE WORK, THE WORKFLOW, AND THE TERMS Deliverables for episode 1 The 6:30 to 7:00 master episode, 4K horizontal, delivered as both a review cut and a final export, plus the project file. Three to four vertical shorts (9:16, under 60 seconds each) cut from the master timeline: the cold open as a standalone short, the probe/Pi story, the lamp origin beat, and the countdown/invitation. Shorts get burned-in captions, styled simply and consistently (most vertical viewing is muted). A thumbnail frame export or two if a great still surfaces while you’re in the footage (not a design task, just an eye-out). Workflow Kickoff: you get the full package (script, footage, assets, scratch VO, references) in one organized delivery, plus a 30-minute kickoff call if useful. I’m in US Eastern time; async works fine and I answer notes quickly. First cut: due within one week of kickoff. I expect the first cut to follow the edit script closely. Where you deviated because you found something better, mark it, because I want to see those. Notes: I give timecoded notes in one consolidated pass, usually within 24 hours of delivery. I’ve edited enough myself to give notes like “hold the tree punch-in 10 more frames and kill the music 200ms earlier” rather than “make it pop.” Revisions: the quote should include two revision passes on the master and one on the shorts. Historically my notes get shorter each pass because the script does the alignment work up front. Final: locked picture, then I record final VO to your timing, you swap it in, we do one last polish pass on the mix, and export. Rates and the test project Budgeted at $15 to 25 per hour depending on experience, or an equivalent fixed price per episode if you prefer (tell me your number for the scope above). I expect episode 1 plus shorts to run 12 to 20 hours for an editor working from a script this complete. Before awarding the full episode, I’m running a short paid test: $50 to 75 fixed for a cut of the 60-second cold open plus title beat from the actual assets. Two or three finalists get the test. It’s real material, you keep the payment regardless, and it tells us both in one hour whether this collaboration sings. If episode 1 goes well: weekly episodes through October at minimum, roughly 8 to 15 hours per week, with the possibility of continuing after launch. I would much rather find one editor who owns this series than re-brief someone every month. The person who cuts episode 1 well gets first right of refusal on everything after it. PART 6: WHO I’M LOOKING FOR The profile You’ve cut narrated storytelling before, and you can prove it with work where the pacing decisions were yours. YouTube documentary style, maker content, video essays, or brand films with a real narrative spine all count. Wedding montages and gaming highlight reels, on their own, don’t. You cut to music natively. When you watch a montage, you feel where the beats land. When a cut is 4 frames late against the track, it physically bothers you. You choose or place the track before fine-cutting, not after. You care about sound as much as picture. The cold open of this episode is thirteen seconds of foley and voice with no music. If that sentence excites you rather than worries you, we’ll get along. You layer room tone without being asked, you notice when a scene needs one wood-click to feel finished, and you mix VO so it sits warm and close. You can follow a script and push back on it. This project comes with an unusually complete edit script. The wrong editor treats it as either a prison (follows it joylessly, flags nothing) or a suggestion (ignores it and improvises). The right editor cuts what’s written, and when they see a better idea, cuts that too and sends both with a one-line note. That’s the collaboration I’m paying for. You communicate in specifics. “The settle feels late, I moved it 20 frames earlier, see v2” is the house style. Vague enthusiasm and vague concerns are equally unhelpful. You’re reliable on a weekly cadence. This series posts on a schedule that’s ultimately set by winter, which does not grant extensions. I need someone who hits agreed dates or renegotiates them before the deadline, never after. Not required You don’t need to know anything about woodstoves, hardware, or startups. The script carries the domain knowledge. You don’t need broadcast credits or a famous client list. You don’t need to be in my time zone; I work async happily, and if you’re in the Philippines or elsewhere in Asia, your evening lining up with my morning actually makes turnarounds faster. You don’t need motion graphics beyond simple, tasteful text animation (the comment cards are pre-built, the title is set type over footage). Who this is not for If your instinct is to add stock b-roll, AI imagery, whoosh transitions, or trending-audio energy to fill gaps, this isn’t your project. There are no gaps. If you need a client to be vague so you have total freedom, this isn’t your project either. The freedom here lives inside the frame: rhythm, feel, sound, the ten thousand small decisions that make the difference between footage and a film. That’s the craft I’m hiring. PART 7: HOW TO APPLY Send me the following, and nothing else: One or two links to narrated storytelling work you personally edited. For your best link, tell me two specific editing decisions you made in it and why. Not a task list. Decisions. A few sentences on your montage workflow: how do you cut to music, mechanically? Walk me through it. Have you worked from a paper edit or edit script before? What do you do when you think the script is wrong? A sound design detail you added to a past project that most viewers never consciously noticed. Your weekly availability for the next month, your committed turnaround for a first cut of this episode with all assets provided, and confirmation of who personally does the editing. Your rate: hourly, or fixed for the episode 1 scope (master plus 3 to 4 shorts, two revision passes). Begin your application with the word “woodstove” so I know you read this brief. Applications that skip it get archived unread, which is the fastest screening question ever devised. I’ll shortlist within a few days, run the paid test with finalists, and award the episode the same week. The footage is shot, the script is locked, and winter is coming whether we’re ready or not. Thanks for reading all of this. If you’re the editor I described, you already know it, and I’m looking forward to seeing your work