Descript

Everyone has one move in their workflow they never explain. The cut that lets an interview breathe. The reason you record the voiceover last. The thing you do to every thumbnail that nobody asks about and everybody notices. It lives in your head. You rebuild it from scratch every project. Next week we're giving you a reason to get it out, with Contra. More details on Monday.

New! Captions that earn their screen time. 18 new caption designs, now in Descript—sorted by theme, from clean and quiet to loud and fun. Scroll to the look you want. Watch the new captions in our video, then add them to yours: https://bit.ly/45TWOxq

You could stop doing all five of these by hand by Friday (that's tomorrow): → sales call → sizzle reel → webinar → filler-word-free cut → one video → ten languages → blog post → video → raw footage → social clips Ask for it in Claude or ChatGPT. Descript does the rest, no window switching. Sign up at https://bit.ly/4gczctO

Most AI labs build general-purpose generative models. We don't. We take hard, specific problems in the editing workflow and build models that solve them. We're hiring researchers to do more of it. Some of what we've worked on lately: - Fix a wrong word without re-recording it. We trained our own continuous neural audio codec and a tightly coupled flow-matching generator for zero-shot speech regeneration. - Audio-driven lip sync. When you edit audio, the speaker's lips stop matching the words. We trained a codec + flow-matching model that regenerates only the facial regions that need it, keeping identity, lighting, and temporal consistency intact. - Jump cut smoothing. Cut part of a video and you usually get an obvious visual jump. Our model synthesizes a physically plausible bridge between the remaining clips, so the edit reads as one continuous take. That's some of the recent work. What's next is the bigger swing: multimodal understanding and agentic video editing—models that watch, listen, reason about, and eventually edit long-form media on their own. That's the work we're hiring for. The team's research: https://lnkd.in/e---QDEQ Apply: https://lnkd.in/eEe7Kfzd

You don't micromanage your video edits anymore. You just ask. Underlord does the work and sweats the details—like catching that B-roll was landscape in a portrait video, and reshaping it before you notice. Reach it from your assistant (Claude, ChatGPT) or build it into your own workflow. No timeline edits required. Sign up at https://bit.ly/4gczctO

Hit Record: The next creator Somewhere in your drafts there's a recording. Maybe you haven't finished it. Maybe you haven't started. Hitting record takes a second. What comes after is where your whole week goes. Audio in one tool, captions in another, an edit that's kind of what you wanted by Friday. June 25, we're doing that live. Raw footage in, finished video out. You'll see exactly what changes and what doesn't. You'll walk away with: ✅ A real workflow, on footage that looks like yours ✅ Live answers from the people who built it ✅ A clear sense of which jobs to hand to Descript and which to keep doing yourself We know what you meant. June 25, we prove it. The hands-on workshop follows on July 1.

Come join my team at Descript! I’m hiring a Content Strategist and Social Media Manager, remote or in SF. We’re a big on the mindset: do the strategy well so you can do the creative part with freedom and earnestness. Looking for a Social Media Manager who has a great sense for what works in spaces like this, knows how to scale employee advocacy, and isn’t afraid to be on camera (+ help others join them!). Likewise, seeking a Content Strategist who can find the story in any data or product doc, has strong ability to meet a variety of audiences, and probably loves what people used to call “writing”. Those don’t fit? Our marketing team is also hiring a Head of Enterprise Marketing, Field Marketing Manager, and PMM. A couple more roles are in the works, so reach out if you market things for a living, like a challenge, or have scathing feedback about Descript and working here would be the only way to convey the true weight of your frustration (I won’t give you a job but I will listen and nod.) descript.com/careers

Anthropic shipped Claude Opus 5 this morning. It's already live in Descript. You can pick it in Underlord and put the newest model to work on the edit you were going to redo anyway. Same co-editor. Newer brain. Available now.

Beauty filters have a real problem: Crank one up and everyone looks like the same airbrushed mannequin. Fine for a selfie. Not for a sales call, a webinar, or a talking head you're putting your name on. Skin smoothing in Descript has sliders instead of one blunt filter. Soften the under eyes, ease the forehead, leave the laugh lines where they are, you set how far it goes, region by region. You record normally, then add it in the edit. No makeup, no perfect lighting, no reshoot. Press play to see it in action.