All The Amazing Things

Brand Design (2025–Present)

Live Links

Overview

My interest in AI vision, generative imagery, and emerging AI visual search for shopping led me to explore how these systems intersect in live commerce environments and real workflows.

I chose interiors as the testing ground — both because it’s a visual domain I’ve long explored and because it demands material accuracy, stylistic coherence, and structured classification. That combination makes it an ideal stress test for evaluating the intersection of product imagery, AI-generated imagery, AI visual search precision, and real-world contexts.

It required defining and extending a disciplined design language across publishing and commerce.

The brand operates across YouTube, WordPress, Pinterest, TikTok, and the Amazon Influencer platform, maintaining a consistent aesthetic framework while adapting to different surfaces, evolving tools, and distribution formats.

It functions both as a public-facing publishing channel and as an ongoing exploration of the intersection of generative image and video tools, emerging visual search capabilities, AI vision-assisted shopping tools, and the preservation of aesthetic authorship and quality as technologies shift.

To complete the immersive experience, all music featured across the video content is entirely AI-generated, adding an additional layer of generative exploration to the brand.

Playlists

Timeless Interiors

A curated lookbook exploring classic design elements, material accuracy, and enduring interior aesthetics.

View the full Timeless Interiors playlist on YouTube ↗

Role

Senior Designer / Brand Owner

Responsible for:

• Defining interior style categories and visual taxonomy
• Establishing material, lighting, and compositional frameworks
• Directing generative and non-generative image, video, and AI audio/music outputs
• Structuring editorial sequencing and pacing
• Designing cross-platform packaging standards
• Organizing storefront classification and curated product groupings
• Maintaining disclosure and affiliate compliance

The brand is independently structured, allowing full ownership of both aesthetic direction and system design.

Objective

To build a cohesive visual brand capable of:

• Extending a clear aesthetic language across multiple publishing surfaces
• Integrating generative tools without aesthetic drift
• Translating inspiration into structured, shoppable editorial output
• Maintaining transparency within affiliate environments
• Evolving alongside emerging image, video, and shopping systems

Brand Structure

Each surface serves a defined role:

• YouTube — Editorial lookbooks and short-form discovery
• WordPress — Indexed archive and written context
• Pinterest — Discovery
• TikTok — Short-form experimentation and reach
• Amazon Storefront — Curated design edits and structured product groupings

Each platform extends the same visual categories and naming systems. Commerce follows the visual framework.

Workflow

The workflow is structured and repeatable:

• Define or refine a visual category
• Generate and iterate imagery within defined constraints
• Assemble structured lookbooks
• Package content according to platform standards
• Map visuals to curated product edits
• Review for visual drift and refine taxonomy
• Publish and adapt based on stats

The visual language acts as the governing layer. Tools and platforms may change; the framework remains stable.

System Insight: Dual Visual Systems

AI-generated imagery and visual search systems operate with fundamentally different visual objectives in commerce environments.

Generative models are trained to produce cohesive, photorealistic scenes that emphasize mood, composition, and overall aesthetic continuity—qualities that align with human perception. Visual search systems, by contrast, rely on discrete object clarity, defined boundaries, and standardized product signals to achieve accurate classification.

Bridging this gap requires a more deliberate visual structure—preserving aesthetic integrity while introducing the clarity needed for computational understanding. This distinction informs how I design compositions, categorize visuals, and translate editorial imagery into shoppable formats.

Tools & Technologies

• Krea AI — For high-fidelity upscaling, enhancement, and generative detailing.
• LoRAs (Low-Rank Adaptations) — For stylistic fine-tuning and maintaining strict aesthetic consistency across batches of generations.
• Meta AI & Google Gemini — For conceptualization, visual search testing, and drafting categorical taxonomy.
• ChatGPT — For prompt engineering, sketching ideas.
• Claude — For advanced contextual reasoning, refining visual classification logic, and structuring detailed editorial frameworks.
• Perplexity — For deep-dive AI research, market trend analysis, and synthesizing accurate information for editorial content.
• Producer AI — For generating original music and full compositions to complete the immersive video experience.
• Canva — For cross-platform graphic packaging, layout design, and final asset assembly.
• Visual Studio Code — For front-end development, custom HTML/CSS authoring, and managing the brand's web architecture.
• Chrome Dev Tools — For live debugging, responsive design testing, and performance profiling of the brand's web surfaces.
• Adobe Illustrator — For precise vector graphics, typography, logo design, and foundational brand asset creation.
• WordPress — For building the editorial blog, managing written context, and structuring the visual archive.
• Amazon Product Advertising API — For programmatic commerce integration, structuring curated product edits, and ensuring affiliate transparency.
• ShopMy — For curating premium shoppable product edits, managing affiliate partnerships, and seamlessly monetizing visual design inspiration.
• YouTube Studio — For managing video distribution, analyzing audience retention, and optimizing editorial lookbook performance.
• TikTok Studio — For publishing short-form video content, tracking engagement metrics, and optimizing reach for discovery.
• RunwayML — For publishing custom animated chatbot AI charcters.

What This Demonstrates

• Ability to define and extend a cohesive brand language
• Structured cross-platform publishing discipline
• Applied taxonomy and classification thinking
• Integration of evolving generative tools without aesthetic compromise
• Responsible alignment of commerce and disclosure
• Maintaining equilibrium between aesthetic discipline and platform velocity

Live Links

Sandbox: Fashion & Wearables

While the core brand strictly focuses on interiors and environmental design, my generative explorations occasionally cast a wider net.

The following collection represents an ongoing sandbox for AI-generated fashion and editorial apparel. Though intentionally kept separate to maintain categorical discipline within the main brand, it serves as a valuable testing ground for fabric simulation, dynamic motion, and future expansion.

View the full unlisted Fashion with AI playlist on YouTube ↗

Current Exploration: AI Persona & Hosting

As the brand evolves, a current design dilemma I am actively exploring is the integration of a persistent AI persona. I have fully developed and fine-tuned a character named Astrid. While I've grown quite attached to her as a creative asset, I am still defining her exact "job" and presence within the channel's established ecosystem.

An Animated Character Featured in Content Videos

This video serves as a quick initial test—an introduction where Astrid welcomes viewers to the channel. It acts as a sandbox to evaluate lip-sync, micro-expressions, and the overall viability of a virtual host interacting with the content.

An Animated Character You Can Chat With

Astrid currently functions as an interactive, animated AI character. While the chatbot is still in beta, she is fully expressive—capable of seeing your screen, speaking, and listening. Her behavior is guided by a customized knowledgebase and strategic scripting.

The Astrid Integration: From Static to Interactive

To create a more immersive user experience, I developed a live conversational interface featuring the brand's custom AI persona, Astrid. This integration successfully merges consistent visual character design with interactive technology.

• Dynamic Engagement: Transforms the site into a two-way touchpoint, allowing users to interact directly with the brand's design perspective.
• Risk Management & Brand Safety: Currently testing in a private sandbox to understand the conversational guardrails and technical requirements needed for a future public release.
• Protecting the Aesthetic: Ensures all AI outputs remain strictly on-brand, maintaining the platform's highly curated atmosphere while preventing unpredictable interactions.

Closing

All The Amazing Things is a brand.

It publishes like a traditional channel.

It evolves with AI creative technology like a design practice.

It is built for change.

The work demonstrates how aesthetic integrity can persist across shifting tools, formats, and shopping behaviors — while remaining structured, transparent, and visually disciplined.