Superalgos Delivers Experimental Integration of TensorFlow

The integration kick starts the platform’s Machine Learning era, leveraging the built-in data-mining capabilities and expanding the infrastructure.

Julian Molina
Superalgos | Algorithmic Trading
3 min readApr 5, 2021


TensorFlow Integration

Beta 9 and the integration of TensorFlow marks the beginning of the AI era in Superalgos. However, the current status of the integration is partial, incomplete, and of an experimental nature. Such an early-stage release goes in line with the project’s agile style of development, seeking early-stage feedback to incorporate it in the following iterations.

The current implementation explores Supervised Learning algorithms, defining the Artificial Neural Network with the typical hierarchical structure of nodes that characterizes Superalgos’ data structures.

If you have experience with TensorFlow, your help in testing the functionality will be greatly appreciated! Getting involved and helping the project advance along this avenue is regarded as a contribution that earns you Superalgos (SA) tokens!

A Supervised Learning algorithm and its Artificial Neural Network definition in the Design Space.

Accessing Multiple Markets and Multiple Exchanges

Another major infrastructure upgrade in Beta 9 allows Trading Systems to access data corresponding to markets and even exchanges different from the default market and exchange.

In Superalgos, any given Trading Session runs on the Market and Crypto Exchanges referenced by the Exchange Trading Tasks and Market Trading Tasks nodes in the Network hierarchy. These are the Default Market and the Default Crypto Exchange.

In addition to the Chart Data Structure object you already know, Beta 9 makes available two new objects: Market Data Structure and Exchange Data Structure.


Let’s say your trading session runs on Coinbase Pro, in the BTC/USDC market.

You suspect that Binance is leading the market, and that ETH/BTC, in particular, may affect capital flows. So, you wish to monitor the current support levels using the Resistances and Support indicator in the Masters Data Mine in the Binance BTC/ETH market.

The following syntax allows you to do just that:

Search the Docs for the “LF Trading Bot Available Data Structures” topic to get all the details!

Bug Fixes, Improvements, and New Tutorials

In addition to the new experimental ML functionality, Beta 9 is packed with improvements iterating on existing functionality, a few bug fixes, and new and improved educational material in the form of new in-app interactive tutorials.

We highly recommend existing users to do the new Basic Education Tutorial in the Getting Started Plugin Workspace. It complements the Welcome to Superalgos Tutorial covering many important topics.

Upgrading From Beta 8

Beta 9 required some refactoring to adjust certain node names. Please, go through the corresponding notes on the README file before upgrading. This will ensure that you can seamlessly migrate your workspaces, strategies, and data to Beta 9.

Token Distribution

The distribution of the Superalgos (SA) token among contributors of the first ten releases — starting with the initial MVP released in October 2018 until the Beta 8 version released in January 2021 — is well underway.

The Superalgos (SA) token is distributed exclusively among contributors. There are plenty of opportunities to contribute to the project, as it doesn’t take much to participate in the collective endeavor!

Sharing whatever you build for yourself — a trading system, an indicator, a tutorial, documentation, a translation, icons, etc. — is deemed a contribution worthy of incentives! Even participating in the Community groups and helping with the onboarding of new users can earn you tokens!

The tokens allocated to each release are distributed among contributors after the release is delivered. The distribution mechanism guarantees that each token in the ecosystem is backed by tangible value added to the project and the intelligence of the collective.

If you haven’t yet!, join the Superalgos Community!

Scammers on Telegram

We are getting an unprecedented number of reports of scammers impersonating group Admins, Core Team Members, and Community Mods. They contact group members in private with all sorts of propositions.

Please beware! People working with the project will never contact you directly unless you contact them first. If a Telegram profile featuring the picture of an admin contacts you in private, it’s a scam.

Be aware of our Community Safety Policy and stay safe!



Julian Molina
Superalgos | Algorithmic Trading

I’m a lifelong entrepreneur and co-founder of, a Bitcoin-inspired open-source project crowdsourcing superpowers for retail traders.