27.jpg

Designing an AI Ecosystem at Sahara AI

Designing AI Products, Systems & Ecosystems

Design Lead

Product Strategy · Product Design · AI UX · Developer Experience · Design Systems · Brand · Design Leadership

Overview

From designing products to designing an ecosystem

Sahara AI was evolving from an early-stage AI product into a broader ecosystem connecting AI, data, developers, contributors, and users, with blockchain infrastructure supporting ownership, attribution, and traceability.

As the ecosystem expanded, the design challenge changed.

It was no longer simply:

How do we design an AI product?

It became:

How do we create a coherent experience across multiple AI products, audiences, and layers of infrastructure?

As Design Lead, I worked across product strategy, product design, design systems, brand, and team leadership, helping the organization move from individual product experiences toward a scalable design ecosystem.

My Role

Leading design across the ecosystem

The Context

From an AI Product to an AI Ecosystem

The Product Ecosystem

 

The Design Challenge

How do we turn a complex AI ecosystem into a coherent product experience?

As the product ecosystem expanded, three design challenges became increasingly important.

01 — Fragmented Products

Different products were evolving at different stages.

This created inconsistencies across:

  • Information architecture

  • Navigation

  • Components

  • Interaction patterns

  • Visual language

Users moving between products could experience them as separate tools rather than parts of the same ecosystem.

Design question

How might we create a shared experience without forcing every product into the same mold?

02 — Increasing Complexity

The products involved increasingly complex concepts:

AI · Data · APIs · Agents · Projects · Tasks · Contributors · Blockchain

The solution was not simply to remove complexity.

Some complexity was inherent to the product.

Design question

How might we make complex systems understandable without oversimplifying them?

03 — Invisible Infrastructure

Blockchain was an important part of the underlying infrastructure.

But most users did not need to understand:

Transactions · Smart Contracts · Hashes · Gas · Blockchain Architecture

They needed to understand:

  • What did I contribute?

  • Where did it come from?

  • Was it recorded?

  • Can it be verified?

  • Can it be traced?

  • What is my role in the ecosystem?

Design question

How do we design blockchain without designing for blockchain?

Design Principles

01 — AI Marketplace

Designing an Economy for AI Assets

From an AI catalog to an AI asset ecosystem

Sahara AI Marketplace was not designed as a traditional marketplace for browsing AI products.

The larger product model connected multiple participants:

Knowledge Providers
provide knowledge and data

Developers
create AI assets

Users
discover, access, license, and use AI assets

Sahara Toolkits
provide the tools for creation

Sahara Blockchain
records ownership, attribution, transactions, and value exchange

This created a fundamentally different marketplace problem:

How do we design an experience that connects AI creation, distribution, consumption, and value attribution into one coherent system?

The Ecosystem Model

Sahara Blockchain

It records key activities across the ecosystem, enabling:

Attribution · Ownership · Verification · Traceability

The Design Challenge

How do we make a multi-sided AI economy understandable?

The challenge was not simply to help users find AI.

It was to make the relationships between:

People × AI Assets × Transactions × Value

clear and intuitive.

My Design Approach

02 — Developer Portal

Designing the Creation Layer of an AI Economy

Sahara AI was not building a traditional developer platform.

The goal was to create an ecosystem where developers could build AI, distribute AI assets, and participate in the value generated by them.

This made the Developer Portal more than an interface to infrastructure.

It became the creation layer connecting developers to the AI economy.

The Strategic Shift

From a Web3 Ecosystem to an AI Ecosystem

The traditional Web3 growth loop was largely driven by infrastructure and transactions:

Infrastructure → Developers → Users → Transactions → Revenue → Incentives → More Developers

Sahara AI was exploring a different loop:

Developers → AI Assets → Consumers → Revenue → Developer Value → More AI Creation

The product question therefore became:

How do we make developers want to build valuable AI, rather than simply use our infrastructure?

The AI Ecosystem Flywheel

 

The key shift was from a traditional Web3 transaction model to an AI-driven creation loop: developers use Sahara infrastructure to create valuable AI assets, consumers use them, and the resulting value feeds back into developers and the ecosystem.

This helped frame the Developer Portal not simply as a developer tool, but as a key entry point into the AI ecosystem—connecting creation, distribution, consumption, and value.

The Design Opportunity

This changed how I thought about the Developer Portal.

A traditional developer portal optimizes for:

Documentation → API → Integration

Sahara's product needed to support a broader journey:

Discover → Build → Create → Release → Distribute → Capture Value

The Developer Portal therefore had to connect creation with what happens after creation.

Product Architecture

The Developer Portal is only the Application Layer

 

The underlying system was structured across four layers:

Application Layer

Sahara Agent · Sahara Toolkits · AI Marketplace · Sahara ID

The layer developers interact with.

Transaction Layer

Sahara Blockchain

Records interactions and transactions.

Data Layer

Data Storage · Data Access

Provides the data required by the ecosystem.

Execution Layer

AI Execution Protocols · AI Execution Abstraction

Handles the execution of AI tasks.

The Design Problem

The architecture was technically complex.

But exposing the architecture directly would make the product harder to use.

A developer shouldn't have to think:

Application → Transaction → Data → Execution

every time they want to build something.

They should be able to think:

What do I want to build?

and let the platform handle the underlying complexity.

This became one of the core design principles:

Design around the developer's intent, not the infrastructure's architecture.

Designing the Developer Journey

From Developer Tool to Creator Platform

This led to an important product distinction.

Traditional Developer Portal

How do I use the API?

Sahara Developer Experience

What can I build, who can use it, and how does it create value?

The second question required the product to connect developer tooling with the broader Marketplace and ecosystem.

Designing the Creation Model

My Design Approach


03 — Data Services Platform

Designing for Enterprise Complexity

A multi-role, workflow-driven data platform

Data Services introduced another level of complexity.

Unlike a consumer product, the platform supported multiple roles working on interconnected processes.

Different Users. One System.

The Design Challenge

Different users interacted with the same underlying platform, but had very different goals.

This led to a fundamental design principle:

One platform does not mean one experience.

The experience needed to provide a shared system while adapting the interface, information, and actions to each role.

Make the System Visible

Complex systems can feel confusing when users cannot tell what is happening.

The interface therefore needed to make visible:

  • Current status

  • Progress

  • Ownership

  • Dependencies

  • Next actions

  • Data state

Progressive Disclosure

The platform contained complex capabilities, but exposing everything at once would increase cognitive load.

I used progressive disclosure to reveal complexity at the moment users needed it.

Complexity still exists. It just doesn't need to appear all at once.

04 — Design System

Designing the Infrastructure Behind the Products

As the number of products increased, solving individual UI inconsistencies one screen at a time was no longer scalable.

The organization needed shared design infrastructure.

The Design System

The system was structured across three levels.

One System. Multiple Product Types.

The Design System supported fundamentally different experiences:

Consumer

AI Marketplace

Developer

Developer Portal

Enterprise

Data Services

The goal was not to make every product look identical.

The goal was to make different products behave coherently.

Design System as Infrastructure

The system became a shared layer connecting:

Product

↕️

Design System

↕️

Engineering

It provided a common language across design and engineering while allowing individual products to maintain their own needs and contexts.

From Components to Patterns

A scalable Design System cannot stop at UI components.

Reusable product logic was equally important.

For example:

Navigation

Onboarding

Search

Data Management

Status

AI Interaction

These patterns could be reused across products without recreating the experience from scratch.

Governance

I also helped establish principles around:

  • Design quality

  • Design reviews

  • Pattern consistency

  • Documentation

  • Cross-product adoption

  • Design–Engineering collaboration

The objective was to make the Design System a living product infrastructure rather than a static component library.



05 — Brand & Product Experience

Connecting Brand to Product

A user's relationship with Sahara AI did not begin when they opened a product.

It could begin with:

Website

Campaign

Social Media

Event

Product Launch

Landing Page

That meant brand and product could not be treated as completely separate experiences.

Design Scope

My work across brand included:

  • Website

  • Product launches

  • Campaigns

  • Landing pages

  • Social media

  • Product storytelling

  • Visual identity

The goal was to create continuity between how Sahara AI was communicated and how it was experienced.

06 — Design Leadership

Building a Design Organization That Scales

As Sahara AI grew, my role evolved from individual product design toward design leadership.

I led an 8-person design team supporting:

Product

Platform

Brand

Marketing

Direction

I helped the team move beyond:

What screens do we need to design?

toward:

Why are we building this product, and what problem are we solving?

Quality

I established a broader definition of design quality across:

  • Product logic

  • User experience

  • Interaction

  • Visual quality

  • Scalability

A visually polished interface was not enough if the underlying product logic was unclear.

Collaboration

I worked across:

Product

Engineering

Business

Operations

Marketing

The role of design was to connect these functions rather than operate as a downstream execution team.

Design Review

Design reviews focused on five dimensions.

Problem

Are we solving the right problem?

Experience

Can users understand what is happening?

Logic

Does the product model make sense?

Scalability

Can the solution evolve with the product?

Quality

Does it meet the design standard?

Growing Product Thinking

A major part of my leadership role was helping designers move from:

UI / UX Execution

to:

Product Thinking

I encouraged the team to ask:

  • What problem are we solving?

  • Who are we solving it for?

  • Why does it matter?

  • Can it scale?

  • How does it connect to the ecosystem?

What I Built

A connected design ecosystem

My work at Sahara AI was not limited to individual product interfaces.

I worked across multiple layers of the organization.

Products

AI Marketplace

Discover AI.

Developer Portal

Build with AI.

Data Services

Create and contribute data.

Infrastructure

Blockchain

Record · Verify · Attribute · Trace

Design System

Shared language · Patterns · Infrastructure

Brand

Website · Campaigns · Launches · Social · Storytelling

From awareness to adoption.

Organization

8-person design team

Direction · Quality · Collaboration · Product Thinking