> Markdown version of [/videos/253-data-fabric-in-action-how-to-enhance-a-stock-trading-app-with-ml-and-data-virtualization?t=3](https://www.wearedevelopers.com/videos/253-data-fabric-in-action-how-to-enhance-a-stock-trading-app-with-ml-and-data-virtualization?t=3). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Data Fabric in Action - How to enhance a Stock Trading App with ML and Data Virtualization

Tired of complex ETL pipelines stalling your ML projects? Discover how data virtualization within a data fabric architecture predicts customer churn without duplicating a single record.

- **Speakers:** Andreas Christian
- **Event:** WeAreDevelopers LIVE
- **Published:** September 22, 2021
- **Duration:** 45:58
- **URL:** https://www.wearedevelopers.com/videos/253-data-fabric-in-action-how-to-enhance-a-stock-trading-app-with-ml-and-data-virtualization

## Summary

Building and deploying machine learning applications within enterprise environments often stalls due to fragmented data ecosystems, opaque metadata, and inconsistent data quality across disparate teams. To solve these distributed data challenges, a data fabric architecture provides a centralized, hybrid multi-cloud integration layer that orchestrates disparate sources without requiring physical data replication or complex ETL pipelines. The IBM Cloud Pak for Data platform, built on Red Hat OpenShift, operationalizes this concept by unifying data engineering, AI governance, and data science workflows under a single control plane. A defining capability of this approach is data virtualization, which empowers developers to query highly heterogeneous databases—spanning NoSQL collections like MongoDB to relational systems like Db2—using standard SQL queries in real-time, eliminating the need to duplicate records.<br><br>Applying these concepts to a stock trading application demonstrates how teams can dynamically predict customer churn by connecting operational dashboards directly to virtualized relational and structured business datasets. Beyond simply making data technically accessible, enterprise platforms must inject context; auto-cataloging features automatically assign business terms and data privacy masks, bridging the gap between obscure technical column names and actual business meaning. Furthermore, integrating tools like Auto AI accelerates the machine learning lifecycle by automating data splitting, algorithm selection, and hyperparameter tuning. Instead of relying on manual trial and error, workflows generate a leaderboard of predictive models, allowing developers to seamlessly deploy the most accurate champion model as an accessible REST API or pipeline. By emphasizing a clear separation of governance duties—enabling data stewards, engineers, and scientists to securely collaborate—a data fabric eliminates infrastructure silos and accelerates scalable AI-driven development.

**Keywords:** data fabric architecture, machine learning lifecycle, ibm cloud pak for data, enterprise data virtualization, heterogeneous database integration, distrubuted sql queries, predictive churn modeling, auto ai model selection, hybrid multi-cloud deployments, red hat openshift, metadata auto-cataloging, data governance frameworks, model deployment rest apis, data privacy masking, role-based data workflows

## Chapters

1. **Defining data fabric for modern data integration** (00:03) — Industry definitions shape data fabric as an emerging architecture for orchestrating disparate data sources across various platforms.
1. **Obstacles in developing machine learning applications** (02:39) — Key hurdles include discovering available enterprise data, understanding content semantics, ensuring quality, and addressing model degradation over time.
1. **Connecting data and modern AI with data fabric architecture** (05:39) — The underlying architecture relies on an organized framework to centralize data governance, visualization, and lifecycle analytics.
1. **Segregating platform roles in the machine learning lifecycle** (09:59) — Distinct organizational structures separate raw data engineering and catalog stewardship from specialized operational model development workflows.
1. **Leveraging Red Hat OpenShift for underlying container orchestration** (13:20) — Kubernetes-backed compute clusters dynamically scale host infrastructure resources and manage automated containerized database deployments.
1. **Preventing customer churn with predictive machine learning models** (15:13) — A reference stock trading platform dashboard evaluates retention risk indicators against embedded customer profiles to surface contextual capabilities.
1. **Exploring internal services and the Watson data catalog** (20:34) — A centralized interface exposes data properties, auto-matches context schemas, and records peer application reviews for distributed operational tables.
1. **Selecting and deploying models with automated AI validation** (25:37) — Algorithmic profiling tools compare predictive hyperparameter limits to automatically generate optimum Python notebook scripts or production-ready REST endpoints.
1. **Integrating cloud databases directly into Python application code** (31:26) — Developers natively query logically linked remote tables using uniform SQL connection structures instead of maintaining multiple runtime data adapters.
1. **Mapping external MongoDB collections into queryable visual tables** (33:43) — Complex architectures like nested JSON documents flatten safely into scalable table views mapping directly to downstream visual join conditions.
1. **Sharing refined analytical data assets with collaborative teams** (39:57) — Role-based permissions allow technical operators to persist transformed information layouts directly back into organizational service catalogs.
1. **Dataset sizing thresholds for automated machine learning utilities** (41:46) — The accuracy of automated prediction frameworks directly correlates with adequate statistical density across representative training sets rather than explicit capacity boundaries.
1. **Exposing generated model endpoints to native mobile platforms** (43:32) — Auto-generated REST structures securely process asynchronous prediction requests from decoupled client networks like mobile or frontend interfaces.

## Related Moments

- [Architecting a unified data and machine learning workbench](https://www.wearedevelopers.com/videos/1525-beyond-gpt-building-unified-genai-platforms-for-the-enterprise-of-tomorrow) (from "Beyond GPT: Building Unified GenAI Platforms for the Enterprise of Tomorrow")
- [Exploring the big data and machine learning portfolio](https://www.wearedevelopers.com/videos/37-alibaba-big-data-and-machine-learning-technology) (from "Alibaba Big Data and Machine Learning Technology")
- [Traditional data architecture before Microsoft Fabric](https://www.wearedevelopers.com/videos/1547-data-analytics-with-microsoft-fabric-end-to-end-use-case-with-data-agents) (from "Data Analytics with Microsoft Fabric: End-to-End Use Case with Data Agents")
- [Career evolution in data engineering and AI platforms](https://www.wearedevelopers.com/videos/1209-coffee-with-developers-maria-apazoglou) (from "Coffee with Developers - Maria Apazoglou")
- [Bringing machine learning into application DevOps](https://www.wearedevelopers.com/videos/990-developer-experience-platform-engineering-and-ai-powered-apps) (from "Developer Experience, Platform Engineering and AI powered Apps")
- [Introduction to Microsoft Fabric and data agents](https://www.wearedevelopers.com/videos/1547-data-analytics-with-microsoft-fabric-end-to-end-use-case-with-data-agents) (from "Data Analytics with Microsoft Fabric: End-to-End Use Case with Data Agents")

## Related Articles

- [MLops – Deploying, Maintaining And Evolving Machine Learning Models in Production](https://www.wearedevelopers.com/magazine/115-mlops-deploying-maintaining-and-evolving-machine-learning-models-in-production)
- [Stephan Gillich - Bringing AI Everywhere](https://www.wearedevelopers.com/magazine/489-stephan-gillich-bringing-ai-everywhere)
- [MLOps – What’s the deal behind it?](https://www.wearedevelopers.com/magazine/125-mlops-what-s-the-deal-behind-it)
- [Making Data Warehouses Fast: A Developer’s Story](https://www.wearedevelopers.com/magazine/107-making-data-warehouses-fast-a-developer-s-story)

## Related Jobs

- [Data Scientist](https://www.wearedevelopers.com/jobs/ext/1351648-data-scientist) at **Almedia**
- [Lead Software Engineer - Data Engineering](https://www.wearedevelopers.com/jobs/ext/2000968-lead-software-engineer-data-engineering) at **Dynatrace**
- [Data & Machine Learning Engineer | Hybrid work](https://www.wearedevelopers.com/jobs/ext/431779-data-machine-learning-engineer-hybrid-work) at **SMG Swiss Marketplace Group**
- [Staff, Machine Learning Engineer (L4)](https://www.wearedevelopers.com/jobs/ext/1202639-staff-machine-learning-engineer-l4) at **Twilio**
- [Staff Business Intelligence Engineer](https://www.wearedevelopers.com/jobs/ext/626164-staff-business-intelligence-engineer) at **Twilio**
- [Senior Software Engineer, Data](https://www.wearedevelopers.com/jobs/48273-senior-software-engineer-data) at **Sportradar Media Services GmbH**