Case Studies

aztlán’s got the know-how in creating successful software solutions, and has the experience to help you with your software development needs. Below you’ll be able to read about a selection of case studies, where we helped clients, with their requests and challenges, and show you the solutions aztlán provided to them.

• Cryptocurrency exchange sourcing framework
• Git Large File Storage Solution

Cryptocurrency exchange sourcing framework

Challenge

The client, a cryptocurrency trading firm, wanted to create an internal software tool that was able to source and parse orderbooks and trades of a multitude of cryptocurrency exchanges and products. The data that would be generated by this tool would be used to create machine learning models that would guide their trading strategies. It was requested that the software was capable of being concurrent, extensible, and scalable.

Solution

A concurrent sourcing framework was created to retrieve orderbooks and trades, that is capable to source a multitude of cryptocurrency exchanges concurrently, and in realtime. The framework was set up that it was easy for the company to extend the software with new exchanges, capable of sourcing either data from websockets and/or rest-api’s, and outfitted with status and notification functionality. Additionally, great care was given towards documentation. Since the client, software developers themselves, were able to quickly understand, extend, and contribute to the code base.

Technology stack

Git Large File Storage Solution

Challenge

Version control systems are excellent in the management of collections of information such as code bases. And, with the introduction of Git Large File Storage this can be extended for versioning of large files as well. This is great, and you’ll be able to use git lfs with Github, or one of the other source control platform, However, the files are typically hosted on the servers of one of these providers. Yet, we wanted to be able to host our own git lfs server and use out own storage backends like using Google Cloud Storage, Amazon S3, or our local filesystem. This is especially helpful for the community that typically want to use a revision system, and that want more control over the provenance and management of large files. For instance: digital artists, game developers, machine learning professionals, and data scientist

Solution

Developed as an internal product, a Git LFS Server implementation was created that is able to respond to the Git LFS Client calls and implements the GIT LFS API. It features the possibility to host the files in a block storage solution (e.g. google cloud storage, amazon s3, etc.), and the local file system. It is set up so that new plugins for other file storing solutions can easily be created. With minimal set up, clients can migrate their current or new Git LFS workflow to this system. Among other future features will include on-premise deployment, and file encryption support.

Technology stack