Arabesque Research

Arabesque AI

Insights

Technology


Portfolio Dragon utilises next-generation technology and development principles.

Cloud-native: Portfolio Dragon is entirely hosted on Google’s Cloud Platform (GCP). This allows it to dynamically scale computational power and storage to deliver performance on demand cost-efficient and sustainably.

API-first: Portfolio Dragon’s architecture allows for easy integration with existing applications and ingestion of new data sources. The platform supports a broad range of interfaces driven by its API-first architecture.

Continuous delivery: Arabesque AI embraces modern software development principles, ensuring a single deployed version of Portfolio Dragon in production, with new features undergoing rigorous testing during the build and release process.

Google Cloud Case Study: https://cloud.google.com/customers/arabesque-ai

AI Research


At Arabesque AI, we are developing an AI Alpha Engine, which provides Portfolio Dragon’s active view which ultimately feeds into Portfolio Construction & Optimization.

The AI Alpha Engine ingests the full array of 250 features from the Data Processing Engine and processes these using a combination of machine learning models to generate an expected return forecast for a 1-month window (AI Signal) covering over 25,000 listed equities. The AI Alpha Engine is re-trained every quarter to counter the natural decay that machine learning models experience to ensure that the engine maintains a high state of performance.

Sustainability & Impact


At Arabesque AI, we believe that economic value creation can and should be combined with environmental stewardship, social inclusion, and sound governance. This is because sustainability factors are more than merely extra financial considerations for investors; they form the very foundations of successful markets. Through our platform, we aim to contribute towards a future where capital is directed towards impacting society and our climate in a positive way.

Check out our other research papers and blog posts here >

research