Pexo is an AI video agent that users direct in plain language, describing what they want while the system selects AI video models, manages the production pipeline, and returns a complete video. The product is aimed at founders, marketers, and small creator teams who need launch clips, brand intros, product reveals, or explainers without stitching together a stack of editors, voiceover tools, and generative models. This launch matters now because the number of usable video models has exploded in the past year, and Pexo's pitch is to give one conversational interface that routes the work across them.
Under the hood, Pexo handles the full production pipeline from a natural-language description to a finished, publish-ready video with music, subtitles, and transitions , and auto-routes each shot to the best of more than ten models including Seedance, Kling, Veo, and Sora . Input can be a URL, PDF, image, video, audio clip, or a written idea, after which Pexo drafts a script, scenes, and creative direction with the user. Revisions happen through a comment-style flow where you mark what you want changed and iterate in chat, which is what the launch tweet is gesturing at with the "vibe-code" comparison.
Alongside the hosted product, the team ships an open-source collection of Agent Skills for content creation across images, audio, and video , with focused entry points like a launch-video skill, a startup-video skill, a saas-video skill, and a founder-video skill for solo founders and small teams . That dual distribution, a chat app for non-technical users and skills that plug into Claude Code, Cursor, and similar coding agents, is what makes Pexo worth a look for anyone building video into a product workflow rather than just posting a one-off clip.
Comments (11)
Sign in to join the discussion.
Y
Yuki Ranta22d ago
The 'mark it on the canvas' bit is the actual feature. Everyone else is still making me write a novel to move a logo two pixels left.
C
Chiaki Nomura22d ago
Curious what the underlying model stack looks like for the mark-to-edit flow. Diffusion inpainting on keyframes, or something more temporal?
A
Adeola F.22d ago
Tweet cuts off mid-sentence on the CTA. Nothing says AI video agent like a broken link in the launch post.
P
Prakash Iyer22d ago
ok wait, if this actually messages me back like a friend I want it to also ghost me when I ask for revisions on a Sunday.
L
Lina Voss22d ago
Rewriting your tagline for free: 'Pexo. Your video co-pilot that never sleeps, never sighs.' You can Venmo me later.
M
Marco Caruso22d ago
Vibe-coding for video is a cute frame but the first 3 seconds of your demo need a hook. I bounced before I saw what Pexo actually does.
R
Rasmus Ek22d ago
Every AI video demo looks incredible until you try to change one shot and the whole thing regenerates into a different person.
B
Beatriz Salgado22d ago
7x24 AI partner sounds lovely until you see the GPU bill. What's the pricing shape, per-minute rendered or seat-based?
G
gitasparks22d ago
Any API or webhook when a render completes? Would love to wire this into a pipeline without babysitting a canvas.
T
Terrence Okafor22d ago
Still not convinced normal people want to make videos. Show me the person who wakes up and thinks 'today I chat with an AI to produce a 30 second clip.'
H
Hyunwoo Shin22d ago
Counterpoint: my group chat sends 40 videos a day. That person is everyone under 25.
Work With Us
Book an intro call to explore our suite of services. We might have a waitlist.
Pexo AI
Hi, I'm Pexo. Your 7×24 AI video partner. Let's make something good together.
Hi, I'm Pexo. Your 7×24 AI video partner. Let's make something good together.
Tags
TLVC Rating
Community Rating
No ratings yet
to rate this launch.
Pexo is an AI video agent that users direct in plain language, describing what they want while the system selects AI video models, manages the production pipeline, and returns a complete video. The product is aimed at founders, marketers, and small creator teams who need launch clips, brand intros, product reveals, or explainers without stitching together a stack of editors, voiceover tools, and generative models. This launch matters now because the number of usable video models has exploded in the past year, and Pexo's pitch is to give one conversational interface that routes the work across them. Under the hood, Pexo handles the full production pipeline from a natural-language description to a finished, publish-ready video with music, subtitles, and transitions , and auto-routes each shot to the best of more than ten models including Seedance, Kling, Veo, and Sora . Input can be a URL, PDF, image, video, audio clip, or a written idea, after which Pexo drafts a script, scenes, and creative direction with the user. Revisions happen through a comment-style flow where you mark what you want changed and iterate in chat, which is what the launch tweet is gesturing at with the "vibe-code" comparison. Alongside the hosted product, the team ships an open-source collection of Agent Skills for content creation across images, audio, and video , with focused entry points like a launch-video skill, a startup-video skill, a saas-video skill, and a founder-video skill for solo founders and small teams . That dual distribution, a chat app for non-technical users and skills that plug into Claude Code, Cursor, and similar coding agents, is what makes Pexo worth a look for anyone building video into a product workflow rather than just posting a one-off clip.
Comments (11)
The 'mark it on the canvas' bit is the actual feature. Everyone else is still making me write a novel to move a logo two pixels left.
Curious what the underlying model stack looks like for the mark-to-edit flow. Diffusion inpainting on keyframes, or something more temporal?
Tweet cuts off mid-sentence on the CTA. Nothing says AI video agent like a broken link in the launch post.
ok wait, if this actually messages me back like a friend I want it to also ghost me when I ask for revisions on a Sunday.
Rewriting your tagline for free: 'Pexo. Your video co-pilot that never sleeps, never sighs.' You can Venmo me later.
Vibe-coding for video is a cute frame but the first 3 seconds of your demo need a hook. I bounced before I saw what Pexo actually does.
Every AI video demo looks incredible until you try to change one shot and the whole thing regenerates into a different person.
7x24 AI partner sounds lovely until you see the GPU bill. What's the pricing shape, per-minute rendered or seat-based?
Any API or webhook when a render completes? Would love to wire this into a pipeline without babysitting a canvas.
Still not convinced normal people want to make videos. Show me the person who wakes up and thinks 'today I chat with an AI to produce a 30 second clip.'
Counterpoint: my group chat sends 40 videos a day. That person is everyone under 25.