Projects

Market platform for the people interested in forex, crypto and other markets

Type Web application
TECHNOLOGIES AWS, Axios, Django, Django Rest Framework, Docker, Gitlab CI/CD, Nginx, PostgreSQL, Python, React, React-vitualized (tables), Recharts (charts), Redis, Redux Toolkit, Sass, Typescript
AREAS OF EXPERTISE Finance, Analytics
TEAM 1 DevOps, 1 Front-end Developer, 1 Back-end Developer, 1 UX/UI Designer, 1 QA engineer, 1 Analyst, 1 Project Manager

Crediblock is a market platform for the people interested in forex, crypto and other markets. It is a convenient resource to find actual trades, quotes, exchange rates, last trade for crypto pairs, view charts and so on.

The Platform allows users to get actual market information in a convenient way and customize views according to the user’s goals (hide and unhide certain fields).

Crediblock is meant to attract new customers to LevelX and promote LevelX Finance (iOS) and LevelX Finance (Android) apps.

Challenges
  1. Virtualized tables
    We have to display data in different large tables according to project specific nature. It has an effect on the project performance.
  2. Adaptive design
    As far as the project nature was about representing the large tables, correct work on the screens with the resolution less than FullHD wasn’t by design in the project initially. But as the project developed we understand that the solution is used frequently on the screens with the resolution less than FullHD. It resulted in problems with the content displaying.
  3. Data caching
    It was impractical to store data from the Polygon in an object-relational database (for example, PostgreSQL) - there were too many of them and some of them were not static - expired immediately after several accesses or at a specific time.
Solutions
  1. ISS developers used a react-virtualized library that allows representing the data in tables without the loss of the performance.
  2. Our frontend developers have implemented the responsive design without according Figma design templates for the screens with the resolution less than FullHD.
  3. We decided to use Redis as a NoSQL database with nesting and clearing logic for certain key patterns for certain dates or events.
Machine learning model for predicting startup success
Machine learning model for predicting startup success
A mobile app to help the users to feel calm based on EMDR
A mobile app to help the users to feel calm based on EMDR