Capabilities — AI Product UX
AI Products
People Actually
Understand and Trust.
We design chat assistants, copilots, agent interfaces and AI-powered dashboards that feel clear, honest and useful from the first prompt, built on the product design work we deliver for fintech, e-commerce, government, healthcare and SaaS teams across Saudi Arabia, Egypt and the Gulf.
Fintech · SaaS
Years of fintech, SaaS and enterprise product design
AR · EN
Arabic-first and bilingual RTL interfaces
Web · Mobile
AI experiences designed end to end
AI features are easy to demo and hard to trust in real use.
Adding a model is the easy part. Designing how people ask, read, verify and recover is where most AI products struggle. Without that layer, people try the feature once and go back to the old way.
People don’t know what to ask
A blank prompt box is not onboarding. Without suggestions, examples and clear limits, users guess, get weak answers and give up.
Answers without proof feel risky
When users can’t see sources, confidence or who is accountable, they double-check everything or stop relying on the feature altogether.
One bad answer breaks the flow
Wrong answers, timeouts and out-of-scope requests will happen. Without designed fallback states and a path to a human, a single failure ends the session.
Our Process
From first prompt to production-ready screens.
Five clear steps that keep product, engineering and stakeholders aligned, and test AI behaviour with real people before it ships.
Discover
We map your users, data, model capabilities and business goals through workshops and stakeholder interviews, so we design for what the AI can actually do.
Define
We decide where AI adds value, what it should never do and how it behaves when it is unsure. That becomes the flows, prompts and guardrails we design around.
Design
We design the conversation, interface and every state for web and mobile: onboarding, inputs, answers, citations, confidence, errors and hand-off to a human, in Arabic and English where needed.
04
Test
We put prototypes in front of real users to see where they hesitate, misread or over-trust the AI, then refine before engineering commits.
05
Hand off
We deliver organised Figma files, components, states and documentation your engineers can build from, and stay close while it gets implemented.
What’s Included
Everything your AI product needs to feel clear and trustworthy.
Pick a focused piece or the full scope. Every engagement is shaped around your product, your users and your model.
Chat assistants & copilots
Assistant and copilot experiences that sit naturally inside your product: entry points, prompts, suggestions, answer layouts and follow-up actions.
Conversational & agent interfaces
Multi-step conversations and agents that act on the user’s behalf, with clear progress, confirmation steps and an easy way to stop or undo.
AI-powered dashboards & data tools
Dashboards and analytics tools where AI summarises, explains and highlights what matters, without hiding the underlying data.
Prompt & flow design
The starter prompts, guardrails and conversation flows behind the interface, designed with your team so the AI’s behaviour matches the experience.
Trust & transparency patterns
Confidence indicators, citations and sources, human-in-the-loop review, and well-designed error and fallback states.
AI onboarding
First-run experiences that show what the AI can and can’t do, set expectations and get people to a useful answer quickly.
Usability testing of AI features
Moderated and remote sessions that reveal where people misread, over-trust or abandon your AI features, with prioritised fixes.
Design systems for AI features
Reusable components and states for chat, streaming responses, citations, feedback and loading, so AI features stay consistent as you grow.
Arabic & bilingual (RTL) AI interfaces
Arabic-first and bilingual experiences with proper RTL layouts, mixed-direction content and copy that reads naturally in both languages.
AI UX expert reviews
A structured review of your existing AI feature against usability and trust heuristics, with a clear list of what to fix first.
AI work, grounded in real product delivery.
We have designed conversational and data-driven products alongside the fintech, e-commerce, government, healthcare and SaaS work that makes up most of our portfolio.
Fintech
Years of fintech, SaaS and enterprise product design, plus e-commerce, government and healthcare.
Bot builder
UX/UI design for a bot builder product.
Chatbot
Chatbot and chat widget design for a retail pharmacy.
AI & data
An expert review, plus product and website design, for an AI and data platform.
Also delivered
Data dashboards for a data platform · Fintech wallets and investment apps · E-commerce apps · Government platforms · Healthcare apps · SaaS dashboards · Design systems · Usability testing
Why teams bring us in for AI product UX.
AI changes what a product can do. It doesn’t change what people need: clarity, control and a reason to trust it.
Product fundamentals first
With years of fintech, SaaS and enterprise product design behind us, we treat AI as part of the product, not a layer on top. Flows, states and edge cases get the same rigour as any core feature.
Trust designed in, not patched on
Citations, confidence, human review and fallbacks are part of the first design, not a fix after users complain.
Arabic-first, bilingual by default
We design for Arabic and English users across Saudi Arabia, Egypt and the Gulf, with RTL layouts that hold up when content mixes both languages.
Research and testing built in
We run discovery workshops and usability tests on AI behaviour, so decisions come from how real people use your product, not assumptions.
Engagement Models
Start focused, or go all in.
Choose the way of working that fits your roadmap. We’ll recommend one after a short call.
AI UX expert review
A focused review of an existing AI feature, with prioritised fixes your team can act on straight away. A low-risk way to start.
Project-based design
A defined scope, from discovery to developer handoff, for a new AI product or a major AI feature.
Flexible monthly retainer
Ongoing design capacity on a monthly plan for teams shipping AI features continuously. Adjust the scope as your roadmap changes.
Pair AI UX with the rest of your product.
Most AI engagements connect to at least one of these.
UX Audit
Find the friction in your current product and get a prioritised roadmap before you add AI on top.
Usability Testing
Watch real people use your product, AI features included, and see exactly where they hesitate.
Product Design
End-to-end product design for web and mobile, from first flows to a polished, build-ready interface.
FAQ
Everything You Need to Know
How long does an AI product UX project take?
It depends on scope. An expert review is the quickest way to start, while a new AI product from discovery to handoff takes longer. We agree a clear timeline with you after a short discovery call.
What engagement models do you offer?
Do you design Arabic and bilingual (RTL) AI interfaces?
How do you hand off designs to engineering?
Can you review an AI feature we have already launched?
How do we get started?
The Difference
AI bolted on vs. AI designed in.
What changes when the experience around the model gets the same care as the model itself.
Bolted on
A blank prompt box and no guidance
Answers with no sources or confidence
Dead ends when the model gets it wrong
English-only layouts that break in Arabic
Engineers guessing at loading and error states
Designed in
Guided onboarding and starter prompts
Citations, confidence and human review where it matters
Designed fallbacks and a clear path to a human
Arabic-first, bilingual RTL interfaces
Every state documented and ready to build
Let’s design AI your users will actually trust.
Tell us what you’re building: a new assistant, a copilot inside an existing product, or an AI-powered dashboard. In 30 minutes we’ll look at where design can make the biggest difference and suggest a starting point.