About Us


Deep Learning & Artificial Intelligence Solutions

We develop Deep Learning and Artificial Intelligence solutions for industry by state-of-the-art research developed in house by experts in the field.

We specialise in:

  • Graph Nerual Networks
  • Vision Based Navigation with Convolutional Neural Networks and Vision Transformers
  • Anomaly Detection on Graph Neural Networks and Temporal Convolutional Neural Networks
  • Entropy Encoding through Deep Auto Encoders

Who Are We?

Machina Doctrina is a collective of dedicated researchers and engineers who specialise in applied Deep Learning and Artificial Intelligence.

We strive to deliver cutting edge technology, combined with world class research in an environment that encourages independent thought.

Get to know our founder:

What are we Building


An Ethereum Transaction Graph Neural Network

We're building a graph representation of all transactions using a Graph Neural Network, specifically a Graph Temporal Neural Network (e.g. such as arxiv.org/pdf/2006.10637).

The aim is to classify either a single transactions (or wallet) or a series of transactions based on the relative relations from both past and present interactions represented as a transaction graph that evolves over time from the Ethereum blockchain network. Typically, we divide the application cases in 3 parts, edge classification/prediction (e.g. classify transactions), node classification/prediction (e.g. classifying wallet types/holders), and graph/subgraph classification/prediction (e.g. transaction load, anomalies)

We have quite extensive experience applying these techniques to social networks (predicting future connections via Twitter) and road networks (predicting traffic load in a sector of the network) and believe there is significant value and insight to be added to the Ethereum network. Such use cases:

  • Peer Discovery
  • Network Anomaly detection

Find Out More...

Did You Know?

Image Image

We generated all the images on this website (baring the profile picture) in house using a Deep Diffusion model!

Here are some that didn't quite make the cut, but we still think are cool.