Embedded Files
Data Architect / Insights & Analytics Engineer | Bi & Data Platforms
Selected works
Interactive Live Dashboard - Click to use
District Energy: Intelligence & Reliability Platform
End-to-End Medallion Pipeline (Fabric/PySpark + Power BI)
Tech Stack & Architecture: Microsoft Fabric (OneLake, PySpark, SQL Endpoint), Power BI (Direct Lake & Import Mode Production, DirectQuery Prototype), Medallion Architecture (Star Schema).
Architectural Rationale: The platform was initially prototyped in DirectQuery to evaluate real-time telemetry potential. For the production suite, I transitioned to a high-performance Direct Lake / Import hybrid. This strategic move leverages the VertiPaq in-memory engine to ensure sub-second interactivity for complex ESG and financial DAX measures, effectively bypassing the traditional trade-off between data freshness and analytical performance.
Strategic Impact:
Business Value: Identified 81,744 MWh in recorded efficiency risk, translating technical grid waste into €4,904,640 of avoidable costs.
Operational Excellence: Engineered a "Single Source of Truth" maintaining 102.59% heat self-sufficiency while tracking ESG targets across a diverse energy fleet.
Core Skills: PySpark, SQL, DAX, Fabric ETL, Medallion Architecture, CI/CD with Git Documentation.
Interactive Live Dashboard - Click to use
Nexus: Supply Chain Intelligence Lakehouse
End-to-End Medallion Pipeline (Databricks + dbt + Airflow + Power BI)
Architecture & Governance: Engineered a Medallion Pipeline (Bronze → Silver → Gold) in Databricks Unity Catalog, refactoring 180k+ rows of nested telemetry into an optimized Star Schema to migrate business logic from DAX to the Warehouse.
Analytics Engineering: Leveraged dbt-databricks to implement MD5 surrogate keys and "Inferred Dimensions," ensuring 100% referential integrity and a 40% increase in downstream report performance.
Orchestration & DevOps: Developed a Python-based Apache Airflow DAG for lifecycle management and implemented GitHub Actions CI/CD with encrypted Secret Management to automate code validation and secure cloud authentication.
Strategic BI Layer: Developed an integrated Power BI Executive Dashboard featuring prescriptive financial modeling (Waterfall Profit Bridge) and sub-second "Drill Down" capabilities into granular shipment-item details.
Quality Engineering: Hardened the platform with dbt 2.0 relationship tests and automated schema validation to eliminate data "leakage" and ensure "Zero-Defect" reporting for stakeholders.
BIP HMI Dashboard - Anomally detection
BIP HMI Dashboard - AI Activated
Bioprocess Digital Twin & Predictive Analytics
Transforming Telemetry into Autonomous Intelligence
In biomanufacturing, discovering a deviation post-batch is a million-dollar risk. I developed BIP: a containerized, AI-augmented architecture that transforms static telemetry into a proactive, closed-loop system.
The Innovation
AI Pilot (Autonomous SCADA): A high-frequency FastAPI engine that monitors batch health and auto-triggers RPM adjustments to stabilize Dissolved Oxygen via physics-based coupling.
GxP Governance: A .NET 8 microservice enforcing 21 CFR Part 11 compliance with role-based e-signatures and immutable audit trails.
Predictive HMI: A React/TypeScript dashboard using moving-window linear regression to forecast temperature trends 60 seconds into the future.
The Architecture
Backends & Logic Python 3.12 • FastAPI • .NET 8 • C# • ASP.NET Core • Pandas
Industrial Frontend React 18 • TypeScript • Node 22 • Recharts • Lucide-React • Tailwind
Data & Infrastructure Docker • Docker Compose • SQLite • EF Core • GitHub Actions • Render
Get in touch at: dalbertonyebuchi@gmail.com
Google Sites
Report abuse