Contact person: Web-based Trading Screenshot Transformation Platform

Phone:Show

Email:Show

Location: Kingston, Canada

Budget: Recommended by industry experts

Time to start: As soon as possible

Project description:
"I am building a web-based platform that turns raw trading screenshots into polished, share-ready images in seconds. The flow starts with a robust OCR/AI extraction engine: users will drop PNG or JPG screenshots from exchanges such as Bybit, Binance, Coinbase and the service must reliably pull out ticker symbols, P&L, entry and exit prices, trade duration and precise timestamps—even when the layout varies from one exchange to another.

The goal is to use our library of images, our library of fonts, but create a studio like feel for users.

I have rough website done with Framer, you will have to work alongside and make sure it can be implemented into it.

The parsed data feeds directly into a template generation layer. I will supply baseline artwork, font families and icon sets; your code should inject the extracted numbers and text, adjust colours and typography on the fly and show a live preview so users immediately see how their trades look inside different themes.

On the front end, a lightweight “creative studio” is required. It needs true drag-and-drop controls, font and colour pickers, image replacement, unlimited undo/redo and real-time updates that feel instant on both desktop and mobile.

When a design is finished, users hit export and receive high-resolution output in PNG, JPG or PDF. Social-media-optimised aspect ratios, batch processing and sub-five-second render times are non-negotiable.

Deliverables
1. Server-side OCR/AI module with exchange-agnostic parsing logic
2. Template population microservice connected to an asset library
3. Responsive React (or comparable) studio interface with live preview
4. Export service delivering multi-format, multi-size renders under five seconds
5. End-to-end test suite proving accuracy of data extraction and speed targets

Code quality, performance and maintainability will be the key acceptance criteria. If you have prior experience with tesseract, EasyOCR, OpenCV, TensorFlow, or similar toolchains, please highlight it in your proposal along with live demos or repos that showcase comparable functionality." (client-provided description)