AI Projects

Establishing next-gen workflows and products.

Front Gate Tickets
2024-Current
Lead Product Designer

UX Copywriting Agent

Status: Launched Q3 2025

Headline

Improving copy quality and consistency by using an AI agent to propose, standardize, and localize UX copy across key flows.

Problem

  • Writing UX copy across multiple applications is a big challenge. Different products have different styles for different audiences as well as their own nomenclature and pre-existing terminology.

  • A comprehensive set of UX guidelines is needed to improve the quality and consistency of UX copy across our suite of products.

  • However the more complex the set of guidelines, the more difficult and time consuming it is to adhere to these guidelines, leading teams to reinvent tone, voice, and naming conventions.

Opportunity

  • Centralize UX copy guidelines using best-in-class resources and product patterns.

  • Leverage the strengths of generative AI to evaluate UX copy decisions against a comprehensive set of guidelines and suggest alternatives.

  • Reduce reinvention of tone, voice, and naming conventions.

  • Empower Design, Product and Engineers with feedback on copy quality and consistency.

Design & Development

  • FGT UX Copy Guidelines

    • A unified, enforceable UX copy standard that human writers and the Front Gate Copilot agent can apply across internal, consumer, and client-facing applications.

    • Features a comprehensive set of standards, tailored for the Front Gate suite of products, including Application-Specific guidance for Internal, Fan-facing and Client-facing products.

  • FGT - UX Copy Pal 9000

    • A custom Copilot agent accessible to anyone in the organization, with instructions based off a condensed version of the FGT UX Copy Guidelines

    • Features the ability to upload .PDFs of designs or wireframes for UX Copy Guideline adherence, and offer suggestions for improvements.

Loom Demo

Experimental Design Workflows

Status: In Progress

Headline

Accelerating early-stage discovery by using AI to rapidly generate, adapt, and test interactive product concepts.

Problem

  • Exploring new product directions (like new shopping flows or upsells) traditionally required substantial upfront design and engineering time.

  • This limited the number of concepts we could test with stakeholders and users, slowing learning and increasing the risk of over-investing in weak directions.

  • Maintaining alignment with design system standards across various applications and codebases is a difficult challenge for users and agents.

Opportunity

  • Use AI to rapidly spin up prototypes as “questions, not answers.”

  • Replace slow discovery with fast throwaway builds to test ideas.

  • Shorten feedback loops with stakeholders and users.

  • Establish next-gen design to development workflows

Design & Development

Prototyping Experiments

I designed a series of experiments to evaluate the effectiveness of different applications, models and frameworks for prototyping. Using a scorecard, I’ve documented my findings for others to learn from.

Design System Starter Kit

I created a design system starter kit that is application and framework agnostic that anyone can use to start prototyping or maintain design system alignment for their own projects. The starter kit is just a zip file that can be dropped into any project. It consists of:

  • Prompts

    • Task-oriented set of chain-linked prompts for agents to reduce variability and improve quality of outputs.

    • The prompts can be evoked by ​simply tagging them, or can be repurposed as custom agents, slash commands, or skills.

  • Tokens.css & Guidelines.md​

    • Defines the core design system rules for properties for theming, typography, spacing, color, components, icons, & accessibility.​

  • Component-Index.md & Figma MCP​

    • Used to retrieve specs for components and designs.​

X-Fader Design System

A tokenized agentic-first design system repository utilizing headless components for rapid design to development workflows.

  • Multi-app, multi-theme, multi-brand design system mapped to Tailwind & Shadcn components​

  • Design system repo with a complete set of components​

  • Private registry to import components and themes from our design system repo to provide an upgrade path for legacy apps​

Loom Demo

AI Readiness & SEO

Status: Roadmapped

Headline

Improve FGT’s discoverability and AI-readiness by structuring content and metadata for both search engines and AI assistants.

Problem

  • As fans increasingly discover events and purchase options through search and AI assistants, FGT needs consistent, structured information about events, pricing, and policies.

  • Existing content is often unstructured or buried deep in the conversion funnel, limiting visibility and making it harder for emerging AI experiences to represent FGT accurately.

Opportunity

  • Turn one of our weaknesses into a strength, by setting the industry standard for AI‑readiness in ticketing.​

  • Improve accuracy of AI and traditional search results.​

  • Surface information about flywheel offerings buried deep in the conversion funnel to AI agents and search providers.

Design & Development

  • Research & Analysis

    • Benchmark with Firecrawl AI analyzer and competitor analysis.

      • AI Readiness ~ Competitive Analysis

      • Test search accuracy through a set of heuristics for each competitior.

    • Benchmark B2B, landing page, event details and waitlist to generate a prioritized list of improvements.

  • Build

    • Improve SEO hygiene: semantic HTML, sitemaps, meta tags.

    • Add AI‑focused improvements: llms.txt, structured metadata, clean .md/HTML assets.

    • Implement JSON‑LD Event schema.

    • Publish XML sitemap.

    • Enhance accessibility and heading hierarchy.

  • Evaluate

    • Re-test benchmark and search heuristic analysis

Previous
Previous

Layaway & Installments

Next
Next

FGT Redesign