Anvika Bharadwaj

Seattle- Based Product Designer• AI • Accessibility • Systems Thinking

I design clarity into complex systems, creating adaptive, AI-driven experiences that simplify data-heavy workflows and scale with feedback.

A young woman with long brown hair, wearing a black sleeveless top and layered necklaces, smiling with arms crossed against a dark blue circular background.

Enhancing Colorblind Accessibility in AI Risk Indicators

  • Project Summary:

    • Role: Product Designer

    • Duration: (insert duration, e.g., 2 months)

    • Company Size: (insert size, e.g., 50 employees)

    • Industry: AI / SaaS / Ethical Tech

    Challenge:

    • Risk indicators relied on color-coding, excluding colorblind users and risking misinterpretation in high-stakes AI workflows.

    Actions:

    • Conducted WCAG accessibility and colorblind simulator testing.

    • Benchmarked UI solutions from finance and tech sectors.

    • Iterated on icon, color, and typography strategies; prioritized color contrast and font adjustments to maintain clarity without visual clutter.

    • Balanced accessibility upgrades with strict layout and resource constraints through rapid prototyping and cross-functional alignment.

    Outcome & Impact:

    • Delivered a scalable, user-centered solution enhancing indicator legibility for all users, including those with color vision deficiencies.

    • The approach is adaptable to other AI platforms where clarity and trust are critical.

    Key Design Insight:

    • Prioritizing contrast and hierarchy over added symbols avoids confusion and improves accessibility for diverse users.

Screenshot of a project management software showing results for an employment screening process, with language, date, access code, overall status, and attempts for each applicant.
Open Case Study
Screenshot of a project management or results tracking software interface with tabs labeled 'Details,' 'Team,' 'Participants,' 'Questions,' 'Results,' 'Access Codes,' and 'Communication'; the 'Results' tab is active. The content displays a list of IRN numbers, languages, dates, access codes, overall statuses, and attempts, with colored status indicators like 'HR,' 'PR,' 'AR,' 'DC,' highlighted in red, yellow, gray, green, and purple. The left sidebar has options for 'Dashboard,' 'Projects,' and 'Admin.'

Enhancing Voice AI Experience in Insurance Customer Surveys

  • Project Summary

    Role: Solo UX/UI Designer | Duration: 3.5 weeks | Industry: AI / Voice Analytics

    Challenge: Users struggled with voice capture because phones had to be held too close, leading to low task completion (<70%).

    Key Actions:

    • Designed walkthrough screens educating users on correct phone positioning.

    • Added real-time visual cues and progress indicators to improve guidance and clarity.

    • Revised CTAs for better user understanding.

    Impact: Improved app learnability and user success rates through simplified, clear interactions and enhanced real-time feedback.

Open Case Study
Screenshots of a smartphone app showing a questionnaire setup, with one screen welcoming users and a 'Continue' button, and another screen for an audio check with a smiling man wearing glasses being prompted to speak.
Screenshot of a voicemail transcription interface showing an audio file with dialogue between a caller, agent, and John, discussing a rainstorm, a cyclist, and a car accident.

Enhancing User Experience in Voice Analytics: Deception Detection

  • Role: Part-time UX/UI Designer | Duration: 10 days | Industry: AI / Voice Analytics

    Problem: Analysts struggled with usability in navigating long transcripts and flagged deception segments, impacting efficiency and learnability.

    Solution: Designed audio controls at the top for natural eye flow, introduced separate dual audio tracks, and replaced full transcripts with flagged segments only. Added timestamps, speaker IDs, and an inline commenting feature.

    Impact: Improved task efficiency by aligning design with analyst workflows, reduced cognitive load, and increased stakeholder confidence with a polished, data-driven design.

Open Case Study
A webpage for Mercury Lounge, a nightclub in Tulsa, Oklahoma. The page features a photo grid of nightlife images including a bar, a music performance, and patrons enjoying drinks. The left side provides location details, social media links, seating capacity, and a description of the venue. To the right, there is a June 2022 event calendar with selected dates highlighted.

Redesigning Cadenzo’s Booking Flow to Boost Conversion Rate

  • Role: UX/UI Designer | Duration: 4 weeks | Industry: Entertainment Tech

    Challenge: Low conversion rate with most users leaving without booking due to confusing booking flow and lack of essential information.

    Solution: Conducted usability research and user feedback analysis to identify friction; redesigned calendar, added featured videos, improved info visibility for artists and venues, simplified booking steps.

    Outcome: Significant improvement in conversion rate with ongoing data tracking to validate effectiveness.

Open Case Study

Research Project

A Study of Adaptive Design Principles in Natural Energy Storage

Open Case Study