- hands-on designer, not a PM.
- Exceptional creative taste first, with enough data sense to test UI directions against user drop-off and enough presence to brief the devs without the founder
- 2–3+ years of real design work.
The founder is the product visionary and intends to stay that way. What he lacks is a designer whose taste he can trust without approving every screen. Right now each UI decision routes through him, and that doesn’t scale while the team tests several interface directions for Videoclaw. The core question is how to keep the flexibility of a prompt box while adding enough guided structure, such as model choices and starting points, that new users don’t stall, without the product turning into a rigid “Canva-style” SaaS tool.
Day to day, this person designs and prototypes competing UI directions from scratch. They ship screens to the devs, then benchmark the resulting user flows against drop-off and session data. The single most important spike is creative sensitivity: noticing that a gray is off, a shadow is too heavy, or a texture or duotone would carry the right emotion, and caring how people feel when they use the product. Data-driven iteration matters but is secondary, since the devs share that load. Leadership needs are light (about 3/10): enough to brief the devs and move work forward without the founder in the room. Long term, the founder wants to grow this person into head of product.
What You’ll Own / Do
Design Videoclaw’s core creation flow end to end, balancing prompt-box freedom with guided structure. Rapidly prototype and test multiple UI directions, then pick winners using user flows, session replays and drop-off data. Set and keep the visual and emotional quality bar across the product, so the founder doesn’t have to approve every screen. Brief the full-stack devs directly and iterate with them in a same-day ship cycle. Look outside video software for visual references that make Videoclaw look and feel different from category defaults. Concrete 30/60/90-day outcomes are pending; the founder framed the UI flow problem as the 30/60/90 goal.
Requirements
Technical
Strong visual and interaction design craft, shown in a portfolio of screens designed from scratch, not only vetted or adapted. Product UI/UX experience on a shipped product. Comfort with AI-assisted design workflows (the founder cited Claude connected to Mobbin’s MCP for UI references). Specific tool stack pending.
Execution & Ownership
Can work with the dev team independently and give clear direction without being a manager. Reasonably organized: tracks own work and decisions. Not expected to be a PM. Uses user data and session behavior to check design calls, not just intuition.
Background
At least 2–3 years of real UI/UX or product design work after school. Internships don’t count toward this, and inflated internship scope is a known concern. Real autonomy on past work: owned screens or flows, not just contributed to a large team.
Location & Visa
In-person in San Francisco at least 5 days a week, at a high-intensity startup pace (see Logistics). Work authorization: US citizens and permanent residents; F-1 OPT accepted; existing H-1B or O-1 holders via transfer. New H-1B or O-1 petitions aren’t workable on this timeline. The founder is open to an O-1 path for an exceptional profile, but it’s slow and uncertain.
Nice-to-Have
Designed for a prompt-to-create product: Lovable, Replit, Bolt.new, Emergent (emergent.sh). The founder calls this “bang on.” Canva design experience, which counts in their favor, ideally with a view on where Canva’s model is dated. Video creation experience, or an active creator presence (Instagram, TikTok, YouTube, Reddit). This is the North Star; the founder will pay premium for it. Design school background such as SCAD. It’s a positive signal for work ethic, never a requirement. Interest in growing toward head of product.
Who Will Thrive Here
Someone who feels a personal pull toward video and creative tools and would work on this problem all day because it opens a new dimension of their craft, not because they were told to. They have strong opinions on texture, color and feel, but defer to users when the data disagrees. They’re comfortable with a founder who is far more intuitive than organized, and supply some of the structure themselves. They want an intense, in-person, ship-daily environment, and they know how to rest so they can sustain it.