Judgment
, not generation.

AI generates options. I decide which ones ship, and turn those decisions into a system your team can reuse.

See how I work
The process

From context to system in five steps.

Each step has one job. It keeps the speed of AI without losing the coherence of the product.

  1. Listen.

    I map users, flows, data and business logic before I design anything. Judgment without context is just opinion.

    Focused scope · Setup · Embedded
  2. Define.

    I set the criteria a screen must meet before AI proposes anything. Evaluation becomes a check, not a guess.

    Focused scope · Setup · Embedded
  3. Direct.

    I work with whichever tools support an agentic workflow and fit your team. I steer their output toward your product and quality criteria. The tools generate. I decide.

    Setup · EmbeddedSee what tools I work with
  4. Curate.

    I keep only what holds up and structure it into components and patterns. The system grows, without screens piling up.

    Setup · Embedded
  5. Hand over.

    I deliver infrastructure your team builds on without me. Design, development and product share one reference.

    Setup

When to bring me in

Signs your product needs a judgment layer.

Principles

The rules I work by, whatever the tools.

  1. -- 01

    AI-enabled, not AI-replaced.

    AI speeds up the work. Every decision that reaches your product is still made by a person.

  2. -- 02

    Apps over websites.

    I design products people use every day, not pages they visit once.

  3. -- 03

    Systems over screens.

    Every screen becomes part of a system your team can reuse and extend, with or without me.

  4. -- 04

    Simplest viable solution.

    If a simpler approach does the job, that’s the one I recommend. Fewer tools, fewer steps, less to maintain.

  5. -- 05

    Pragmatic, not hype-driven.

    New tools earn a place in my workflow when they prove useful in real work, not when they trend.

The approach in practice

NxtPeople: one system, multiple products.

Superflow built NxtPeople-DS for a consistent HR platform and its GoTemp sub-brand.

NxtPeople product interfaces, including login, AI assistant, mobile planning and scheduling.

The challenge

Multiple HR modules and GoTemp needed one consistent foundation.

The decision

Shared components, with semantic tokens separating brand from function.

The rejected option

Separate libraries per product: more maintenance, less consistency.

The outcome

One system for NxtPeople and GoTemp. New screens reuse shared components.

NxtPeople product interfaces, including login, AI assistant, mobile planning and scheduling.

Approach

What happens in a free sparring session?

In one hour, I go through one flow, screen or workflow question from your product. I give concrete UX and workflow advice, and you get three written takeaways afterwards. You also get an honest view of whether I'm the right fit. If I'm not, I'll tell you. If there's a next step, I propose a scope and budget.

What happens after the sparring session?

If there's a fit, I propose a scope and estimate a budget for it. That can be a focused UX review of your core flows or a full Design-to-Dev Setup. You decide whether to go ahead.

Do you look at our way of working, or only at design?

Both. Most product problems don’t come from bad screens. They come from how design, development and AI-generated output connect, or fail to connect. I talk to founders, product leads and developers, and I review your codebase structure alongside your designs. Good design decisions need that full picture.

What do we get from a Design-to-Dev Setup?

A foundation your team can build on: a design system that serves as the single source of truth for designers, developers and AI agents. I take one real feature through the new workflow end to end. That proves the approach works on your product, not in theory.

How do you work with AI agents in our design-to-dev workflow?

I set agents up to work from your design system: tokens, components and usage rules they can apply consistently. Then I add review gates, so AI-generated design and code are checked against clear quality criteria before they ship. The AI makes you faster. The judgment keeps your product coherent.

Do we need a design system before AI tools are useful?

Not to start, but AI output only stays consistent when tokens and components are defined. Without them, every generated screen drifts a little further.

Which tools do you work with?

I’m tool-agnostic: I pick whatever fits your team and keeps the workflow efficient. I do have a preferred stack. Figma holds the design tokens and system. Figma MCP, Claude Code and Codex are my AI agents. Linear or ClickUp handle project management, and GitHub the code. On the frontend I usually work with Vue or React, often with shadcn/ui components, on a Laravel/PHP or Node backend. If your team already uses something else, I adapt to it rather than the other way around.

See all questions

Start with step one.

Every engagement starts with a free sparring session. If there’s a fit, I propose a scope and estimate a budget for it. You decide whether to go ahead.

Book a free sparring session

Ask Superflow

AI assistant · answers from this site

Superflow AI

Questions about working with Bart? Ask about the approach, services or your next step.

AI can make mistakes. Messages are sent to OpenAI to answer your question. Don’t share sensitive information. Privacy