Singular: from zero to two decks in a week
How I used five generative AI tools in parallel to explore six visual directions — and change the type of conversation with the stakeholder.
The N5 CEO brings down a vision: Singular, a B2B customer engagement platform with AI for banks. There's technical architecture, there's a commercial hypothesis, there's strong intuition about the market position. What there isn't is a designed product, a visual narrative, or a concrete way to show the team and clients what this thing looks like.
The brief was clear: in one week, produce something that worked both for internal alignment and for selling. With no more context than that.
The starting point: what Singular is not. The value prop is understood in a second.
The first thing I did wasn't open Figma. It was understand the product — and the first design decision I made changed everything that came after: Singular isn't one product, it's two.
An operational panel for the bank's CRM Manager — high density, dark mode, real-time actions. And an executive panel for the C-level — read-only, top-down narrative, strategic synthesis. Treating them as one thing would have compromised both. Splitting them was the first deliverable of the week, and it was conceptual: I hadn't drawn anything yet.
From there, the work ran on two tracks in parallel.
The first was the UX deck — the experience architecture of the full system. How Singular works as a product, its three layers (the brain that decides, the voice that executes, the client who lives the outcome), user archetypes, channel logic.
The three layers of the system in Lápiz Style: the brain (HOPE), the voice (North Star), the client. Illustration generated with Midjourney from the system's canonical visual prompt.
"The bank defines the universe. North Star activates the right version." — Slide from the UX deck.
Valentina, North Star's voice. A WhatsApp conversation as a concrete materialization of the end client's experience.
The second track was the visual admin exploration. Instead of proposing one direction and defending it, I decided to map the full space: five different generative AI tools, same brief for all, six visual variants in parallel. Not to show I could produce a lot, but so the CEO could choose with real criteria instead of approving blindly.
The insight that emerged in the process: each AI tool has a "default voice" that pulls toward a style. Lovable pulls toward clean fintech, v0 pulls toward Linear, Stitch pulls toward dense. Fighting that voice wastes time. Deliberately leveraging it — using each tool for what it does best — is what allowed exploring six directions in a day.
| Variant | Visual direction | Tool |
|---|---|---|
| A | MOB Light — N5's own design system, light mode | Claude Design |
| B | MOB Dark — N5's own design system, dark mode | Claude Design |
| C | Liquid Glass — light physics, visionOS applied | Claude Code |
| D | Bloomberg Style — maximum data density | Lovable |
| E | Bugatti — total monochrome, extreme discipline | Stitch |
| F | Lápiz Dark — editorial cream in dark mode | v0 |
For each variant, we documented which tool was used, what prompt it was given, what had to be corrected by hand, and how quickly we could iterate. That documentation is part of the deliverable.
Variant D · Bloomberg Style — Lovable. Maximum data density, no aesthetic concessions. The direction closest to the real banking context.
A · MOB Light — Claude Design. The in-house design system applied to the product. Full coherence with N5 Now.
B · MOB Dark — Claude Design
C · Liquid Glass — Claude Code
E · Bugatti Style — Stitch
F · Lápiz Dark — v0
Two actionable decks at the end of the week. The UX one to align the team on what's being built. The visual one so the CEO could make a real decision on aesthetics and direction — with complete traceability of why each variant was considered and why it was ruled out.
The methodological thesis that guided everything: AI as co-pilot, not executor. Every design decision was mine. Every asset production went through at least one AI tool. The designer's role doesn't disappear — it becomes director, curator, and critic of output.
The CEO walked into the final meeting with a real decision to make — and he made it: Bloomberg Style, the maximum data density direction. The closest to the real banking context, and the hardest to defend without having explored it against the other five. That's what matters most to me about this case: not the volume of what was produced, but the type of conversation it enabled. Instead of "I'm bringing you what I decided," the dynamic was "I'm bringing you the space for you to decide."
And what was demonstrated — at least for me — is that speed used well isn't a shortcut. It's what lets you explore the problem instead of solving just one version of it.
Singular has a product in development. But when the CEO needs to deliver a keynote at Fintech Americas, slides aren't enough. The brief was to make a sharp turn: forget the product for a moment and think like a show. This demo isn't Singular — it's the concept car at the motor show. A functional piece so the room can see where the system is going before it exists.
View live demo →