You are viewing a preview of this job. Log in or register to view more details about this job.

REACT NATIVE Engineer (Tier 1 VC backed startup - With attached Coding Challenge)

Crema Social
Email

Are you a REACT NATIVE engineer? We are looking for someone highly experienced to come live in New York City and be a part of a fast growing social consumer app with 1M+ users, backed by the who's who in Venture Capital (A16Z, DCM). 

 

If you have not developed an APP using react native framework from scratch, your resume won't even be read. We don't care about your references, transcripts, or extras that you may think is important. What matters is whether you read through all of the requirements of his job description and followed them? Sounds simple? Because most of you can't follow it.

 

This is a springboard like no other if you are the star engineer at your college. Must have had built apps with react native in the past and go through a coding challenge to ensure you have the skills to execute. 

 

True Garage Startup mentality, but you'll like and work in our HQ in a high rise luxury building. You'll be exposed to all that Silicon Valley has to offer right in NYC. From this you will be able to write your own ticket having worked with proven entrepreneurs with previous unicorn exits. Not for the faint of heart!

 

Want to stand out? email tim@cremasocial.com and show me your can complete this challenge. You will be considered if you can really execute.

Goal

Create a mobile application in React Native where the user can:

  1. Select or upload an MP4 video.
  2. Detect if the TikTok watermark/logo is present in the video.
  3. If a watermark is detected, blur or remove (inpaint) that region.
  4. Provide:
    • A flag (boolean) indicating if the watermark is detected.
    • A processed video with the TikTok watermark removed (or obscured).

Requirements

React Native Frontend

  • A simple UI to pick an MP4 file from the user’s device.
  • A display of whether a watermark was detected (true/false).
  • If a watermark is detected, show (or provide access to) the processed video.
  • You may use any libraries for video picking (e.g., react-native-document-picker) and video playback (e.g., react-native-video).

Backend Service

  • The video should be sent from the React Native app to a backend for processing.
  • The backend can be built in Node.js or Python (or a language of your choice if you have a strong preference).
  • The backend should:
    1. Receive the uploaded video.
    2. Detect whether the TikTok watermark is present.
      • A simple or heuristic approach is acceptable; advanced detection is a bonus.
    3. Remove or blur the watermark area from the video.
    4. Return both the detection result and a link or direct file for the processed video to the React Native client.

Watermark Removal Approach

  • You can blur the watermark region, use FFmpeg filters, or attempt an OpenCV-based inpainting method.
  • Support for multiple watermarks or moving watermarks is a bonus.

Deliverables

  • A GitHub repository (or similar) containing:
    1. The React Native project.
    2. The backend code (if separate, include instructions on how to set it up and run).
  • A README detailing:
    • How to install and run the app (frontend + backend).
    • Which libraries you used and why.
    • Any trade-offs or assumptions you made.
  • Short demo or screenshots are a plus.

Bonus Points

  • Handle multiple watermarks or watermarks that move around during the video.
  • Handle multiple video resolutions (360p to 1080p).
  • Provide a configurable approach to choose between blurring or inpainting.
  • Include basic tests or scripts verifying the detection and removal process.
  • Optimize for scalability and speed with minimum computational resources.
  • Any creative ways to make this solution better :)

What We’re Looking For

  • Code organization: Clear project structure and modular code.
  • Technical decisions: How you choose to detect the watermark and remove it.
  • Practical implementation: Even if the detection is heuristic, demonstrating the approach is valuable.
  • Documentation: A concise yet informative README and comments.
  • Usability: A simple UI flow to pick/upload a video, show detection result, and display the processed video.