BrandScore AI

• 

  Brand Score

• 

  How it works

• 

  History

• 

  Landing page

Inspiration

In the age of digital presence, our profile pictures and selfies often serve as first impressions whether we're job hunting, networking or building a personal brand. I was inspired by how something as simple as a profile image can influence perception and wanted to create a tool that demystifies this impact using AI. The goal was to build something that is not only insightful but fun, viral and useful for everyone from job seekers to influencers.

What it does

BrandScore AI lets users upload a selfie or social media image and receive a real-time "Brand Score" out of 100. It analyzes factors like facial expression, posture, eye contact, lighting, and background then generates subscores for Confidence, Professionalism, Trustworthiness, and Approachability. The app provides actionable tips to improve your image and includes optional features like saving score history, sharing your score, and seeing how your photo style evolves over time.

How we built it

The web app was made with Bolt.new. The frontend was built with React and styled using Tailwind CSS for a modern, responsive design. MediaPipe FaceMesh runs in the browser to detect facial landmarks without needing to upload images to a server, keeping the experience privacy-first. I used a mix of heuristics and lightweight TensorFlow.js models to compute subscores. User authentication, storage and image history were handled via Supabase.

Challenges we ran into

Ensuring that all image processing remained fast and private using only client-side tools. Designing a fair and explainable scoring system that users can trust and understand. Building a fun and polished UI that works well on both mobile and desktop devices.

Accomplishments that we're proud of

Built an AI tool that feels personal, playful, and useful not just another tech demo. Achieved completely client-side processing, enhancing speed and user privacy. Created a viral-ready experience that encourages users to share and compare scores. Added real value for professionals and creators looking to elevate their digital presence.

What we learned

Client-side computer vision (with MediaPipe) can be surprisingly powerful and accessible. Simpler scoring systems combined with good UX can feel just as magical as deep AI models. People are excited by feedback tools that help them improve their image without judgment.

What's next for BrandScore AI

Add scoring presets for platforms like LinkedIn, Instagram, and dating apps. Offer premium insights including emotion analysis and style breakdowns. Partner with personal branding experts to offer feedback and consultation.

Built With

• bolt.new
• mediapipe
• react
• supabase
• tensorflow.js