top of page
AI Clone Feature
Scaling authentic creator conversations through human-centered AI
3 months
Product Designer
Figma, Google Gemini, Product Research, User Research
Overview
AI Clone is a feature designed for Markit AI that enables creators to engage their audience at scale using a personalized AI assistant trained on their tone, values, and content. The project focused on designing a system that preserves trust, transparency, and creator control while reducing the cognitive and time burden of managing high volumes of messages.
THE CHALLENGE
Problem Statement
Creators want to maintain meaningful connections with followers, but one-to-one communication does not scale. Existing automation and chatbot tools often feel impersonal, opaque, or risky, especially when they blur identity and authorship. The core challenge was to design an AI-powered experience that feels supportive rather than deceptive, and assistive rather than replacement.
DISCOVERY
Research Insights
I conducted competitive analysis across AI character and creator tools and reviewed existing Markit user workflows. Key insights included:
1. Creators are concerned about losing authenticity and voice when using AI
2. Users need clear disclosure that followers are interacting with an AI assistant
3. Onboarding for AI tools often creates cognitive overload and confusion
4. Creators want granular control over what the AI can say, learn, and access
5. Trust increases when the setup is gradual and purpose-driven
APPROACH
Design Process
Discovery & Research
Conducted competitive analysis of AI character and creator tools and reviewed Markit AI’s existing workflows. Analyzed user pain points around authenticity, trust, and scalability. Synthesized findings into key behavioral and ethical constraints for AI assisted communication.
Ideation and Concepting
Explored multiple interaction models for creator-controlled AI assistants through sketching and flow mapping. Defined core design principles, including transparency, creator control, and gradual onboarding. Prioritized concepts that aligned with user mental models and reduced perceived risk.
Prototyping
Created low-fidelity sketches and task flows, followed by high-fidelity wireframes in Figma. Designed a step-by-step onboarding experience that guides creators through defining purpose, tone, expertise, and permissions. Iterated interaction patterns to minimize cognitive load and setup friction.
Usability Testing and Iteration
Reviewed designs with internal stakeholders and the Head of Product to validate clarity and feasibility. Iterated copy, hierarchy, and entry points to reinforce that users are interacting with an AI assistant. Refined progressive settings and keep synced functionality to support long-term usability and trust.
IMPACT
Outcomes & Results
The AI Clone feature provided creators with a scalable way to stay present without burnout while maintaining trust with their audience. The design aligned human factors principles with emerging AI capabilities, positioning Markit AI as a human centered communication platform. This project demonstrated how thoughtful UX and Human Factors-driven decisions can guide ethical and usable AI systems in real-world products.
Reduced creator onboarding friction by breaking AI setup into clearly defined steps, decreasing incomplete setup drop-off observed during internal reviews
Improved clarity and trust in AI interactions through explicit disclosure and assistant framing, reducing user confusion around authorship during design validation
Enabled scalable creator audience engagement by introducing AI-assisted responses that reduced manual message handling effort during pilot use
Improved long term usability by implementing keep synced functionality, minimizing repeated retraining and ongoing maintenance for creators
DESIGN
Sketches & Figma Prototypes
Onboarding
AI Clone Dashboard
bottom of page