We turn long videos into short-form
content at scale.
Zo Clipsis an AI video platform that takes a single long video and returns publish-ready vertical clips — found, cut, captioned, reframed, and scheduled. The product is built and live. We’re raising to expand the inference capacity behind it.
Where we are today
These figures are maintained by the founding team and are shown here exactly as recorded — no modelled or projected numbers are presented as actuals.
One workflow, not a stack of tools
A creator pastes a link, uploads a file, or imports from Google Drive. Zo Clips handles everything from there — the six capabilities below are shipped and running in the product today.
AI clip discovery
Transcribes a long video, then scores and ranks the moments most likely to perform as standalone shorts — replacing the manual scrub-and-cut pass that dominates editing time.
Automatic captions
Word-level captioning with styled, animated caption templates burned into the vertical render — the format short-form platforms reward.
Speaker tracking & reframing
Computer-vision services detect and track faces and subjects, reframing 16:9 footage to 9:16 so the speaker stays centred instead of drifting out of frame.
Multi-language support
Transcription and captioning across languages, widening the addressable market beyond English-only creators.
Publishing & scheduling
Connect social accounts and schedule or publish finished clips directly, closing the loop from source video to posted content.
Analytics
Performance tracking on published clips, feeding back into which moments and formats are actually working.
The timing case
Short-form vertical video is the default distribution format across every major platform, and the volume demanded of each creator has outgrown what manual editing can supply.
The underlying models — speech-to-text, vision, and language — became cheap and good enough to automate the full pipeline only recently; this product would not have been buildable at this cost two years ago.
The market is consolidating around end-to-end tools. Point solutions that only cut, only caption, or only schedule are being replaced by a single workflow, which is the position we are built for.
Why the margin holds
Most tools in this category resell a third-party API and inherit its price floor. Our architecture is a deliberate bet in the other direction.
Self-hosted inference, not per-call resale
Transcription, face detection, subject tracking, and embedding services run as our own services rather than as marked-up third-party API calls. The heavy per-video compute is infrastructure we control, so cost per clip falls as utilisation rises instead of scaling linearly with a vendor's price list.
Bring-your-own-key absorbs the variable cost
Users can connect their own LLM provider key, which moves the most volatile line item — third-party model spend — onto the user during early access. That keeps gross margin defensible while the paid tiers are still being validated, and it is already shipped, not planned.
What we're raising, and what we need beyond capital
Money is the smaller half of the ask. These are the four things that would most change our trajectory.
Capital for inference infrastructure
The single largest constraint today is GPU capacity. Funding converts our compute from a bottleneck that limits how many users we can serve into headroom, and lets us cover model costs for users on paid plans instead of asking them to bring their own key.
Runway to convert free users to paid
The product, billing, and subscription tiers are built and live. What we need is the runway to grow the top of the funnel and validate pricing against real retention, rather than optimising for cash preservation.
Distribution and go-to-market help
Introductions to creator networks, agencies, and media teams who publish at volume. Our best early accounts come from operators who already feel the editing bottleneck daily.
Operating experience in AI infrastructure
Investors who have scaled GPU-backed products and can pressure-test our cost-per-clip model, vendor mix, and hiring plan as usage grows.
Let’s talk
We’re happy to share the full deck, a product walkthrough, and detailed metrics under NDA. Reach out and mention you’re investing — it goes straight to the founding team.
Contact the foundersThis page is for information only and is not an offer to sell or a solicitation of an offer to buy any security. Any investment would be made solely on the basis of formal offering documents.