Edge Intelligence is built with a focus on open standards and ease of integration with client tooling. Our goal is to make the network information collected by Beeks Analytics easily accessible and analysable within the firm (not just by Beeks Analytics dashboard specialists), and to provide AI/ML tooling which will support firms integrating other data alongside the Beeks Analytics data for their own AI/ML modelling.

One of the foundations of Beeks Market Edge Intelligence™ is data agility and openness. To provide full access to analytics data at scale, Market Edge Intelligence™ allows direct database access via AIML-enabled SQL and bulk big data access via Kafka, Apache Arrow, or Parquet.

Edge Intelligence CDF-Q for rich data analysis

QuestDB can be populated by Beeks Analytics (it can act as the timeseries and message store).

QuestDB is an Apache 2.0 open source columnar database that specialises in time series.

In addition, you can also bring your own timeseries data.

(Note that QuestDB can also be directly accessed via SQL queries).

Using Python code, you can easily bring your own models to the platform to supplement the standard models provided as part of the solution. Python can also be used to dynamically query PCAP data.

Beeks Market Edge Intelligence™ provides exogenous data (economic calendar, holidays, weather) to enrich your system performance data.

Beeks Market Edge Intelligence™ leverages MindsDB as a training and exploration harness. MindsDB enables users to create, train, and deploy AI models quickly and efficiently, all within a framework designed to support real-time analytics. The platform allows users to experiment with models and historical data to seek out patterns that could be used to provide insight running in real-time in future.

The MindsDB platform helps developers orchestrate AI/ML-to-data pipelines, connecting AI/ML models with live data. Using MindsDB automation capabilities, you can build custom AI workflows, bridging hundreds of AI/ML models, enterprise data platforms, applications, and vector stores.

By tapping directly into QuestDB’s centralised repository, Beeks Market Edge Intelligence™ has straightforward access to the full spectrum of trading data—from market events and network performance metrics to decoded order messages. This comprehensive visibility allows AI models to detect anomalies, predict trends, and perform complex analytics with greater accuracy and lower latency.

Jupyter Notebook is a tool that allows users to enter code into ‘notebooks’ and view the corresponding outputs such as visualisations, text, or tables. Beeks clients can use our hosted Jupyter Notebooks service to explore edge data by running custom queries against Beeks Market Edge Intelligence™ datasets. Jupyter Notebooks can be used not just for analysing data in QuestDB but also for automating lower level packet capture analysis as well.

Beeks Market Edge Intelligence™ can also populate Beeks' Grafana tool - VMX-Explorer for dashboard visualisations.

The use of QuestDB also unlocks other ways of accessing the data in the system. See Why Beeks Analytics chose QuestDB as our database engine of choice to unlock the value of your wire data in the Analytics Concepts Guide for more background.