Forward Deployed AI Engineer

Solving problems with Data & AI

My story isn't that of a traditional developer. It is the story of someone who learned the business from the inside out, from procurement operations to data engineering and AI architecture.

Discover my story View Projects

The Journey

From corporate inventory to consultative data engineering.

2009

Stockkeeper and Buyer

Where it all began. Inventory management, parts requisition, and strategic purchasing calculating twice the quarterly average to ensure operational efficiency. I learned to deal with the chaos of the shop floor.

Transition

Buyer (MRO & CAPEX)

Here I realized the limits of purely reactive operations ("putting out fires"). I created the company's first consolidated contract control, integrating it into the ERP and generating financial visibility and strategic discounts that no one else saw.

ERP Integration

SAP Planning and Migration

Actively participated in data and product migration. Focus on purchasing planning in SAP, portfolio management and follow-up, thoroughly understanding how large-scale systems operate the supply chain.

The Turning Point

Transition to Data (Logistics)

Embedded in the heart of the business. Without modern ETL tools (low-code), I automated manual SAP ERP reports (inventory, distribution, assets) deciphering complex procedures in Oracle SQL, VBA, and Excel.

Consulting

Data Consultant

Corporate agile environment (Scrum). Data extraction via Azure Data Factory, robust semantic modeling in Azure Analysis Services (AAS) dealing with IoT sensor data and Data Warehouse (Synapse), delivering direct value to the end client.

Current

Data & AI Engineering Consulting

Intense focus on engineering. Consumption of complex data via APIs to SQL, automations via SAP RFC, and resilient architectures. I am certified in Microsoft Fabric and work in building and managing modern Lakehouses. My focus also expands to creating autonomous flows with Agentic AI to solve corporate problems impossible to scale manually.

Success Cases & Experience

Technical details of key delivered projects and architectures.

ERP to DW Migration: Built an Oracle DW (Star Schema) for an ERP migration from Protheus to Oracle Fusion (manufacturing). Acted as the sole consultant responsible for the ETL layer in IBM DataStage (50+ tables, Dev/Prod) for 7 months.
Dashboard Automation (Fabric): Replaced a manual Excel process (10-30 min/hour) with a Lakehouse in Fabric (Bronze/Silver/Gold, 7 tables), configuring Fabric-Oracle EBS connectivity from scratch.
Multinational Financial Lakehouse: Built a Lakehouse (BR/AR/CL/PE) with 20-25 tables (up to 600K rows) and daily pipelines. Delivered in 4 months with zero errors, supporting multiple dashboards.
SAP HANA Cloud Migration: Migrated over 1,000 tables (200M+ rows) to Oracle Exadata due to license expiration, preventing data loss via Python and ABAP scripts.
SAP ERP Transition: Migrated 200M+ rows of legacy financial data from SAP HANA to Microsoft Fabric, preserving the history for the upcoming ERP change.
WhatsApp SaaS Platform (Full-stack): Built a multi-tenant SaaS (Next.js 16, PostgreSQL, Docker) for marketing via Evolution API. Architecture with queues (Node.js) and Python ETL synchronizing Fabric customers.
Chatbots & Agentic AI: Refactored real estate chatbot on AWS (structured LLM + regex fallback); created multi-agent support system (LangGraph, Databricks OAuth, AD) and deployed memory/observability for an agribusiness LlamaIndex bot.
Pipelines & Automations: Python SFTP transfer (Payroll); real-time API ingestion (SQL Server/Jenkins); government RPA (SIGOR) saving weekly hours; and SAP RFC extraction to Delta Tables.
Modeling & Pricing (Databricks): Modeled competitor pricing dashboard (~15M rows) crossing internal and third-party data for ROI analysis. Silver layer transformations refactoring non-native functions for reliability gain.
Semantic Layers (Trino & AAS): Created semantic layers via Trino on Databricks and Azure Analysis Services (Advanced DAX) to feed Power BI models using Figma visual standards.
Data Warehousing (Azure): Pipelines in Azure Data Factory ingesting data (CSV, SQL, Oracle, SAP HANA) into a Data Warehouse (Star/Snowflake) in SQL Server Azure.
Large-scale IoT Monitoring: Technical leadership in IoT sensor monitoring demanded by 40+ stakeholders. Modeled in Oracle delivering a high-usage analytical operational solution.
Mission Critical & RPA: Maintained PL/SQL packages feeding distribution for 250+ resellers. Built daily Python infographics (Pandas, Dash) unifying production/inventory, and RPA/ETL connecting SAP reports to Oracle DW.

Tools & Methods

Technical stack and methodologies applied in practice over the years.

Data Engineering & Cloud

Microsoft Fabric: Lakehouse Architecture (Medallion)
Python: Pandas, Dash, Custom ETL and RPA
Azure: ADF and Synapse Data Warehousing
Databricks (Spark): Silver/Gold transformations
IBM DataStage: Enterprise extraction
AWS: Jenkins, Glue, EC2

Databases & Systems

Oracle: PL/SQL, Exadata, Procedures
SAP HANA & RFC: Massive extraction and migration
SQL Server: Dimensional Modeling (Star Schema)
PostgreSQL: SaaS transactional databases

Artificial Intelligence (Agentic)

LangGraph: Multi-agent flows and orchestration
LlamaIndex: Metrics and observable RAG
LLMs: Structured data extraction
Redis: Conversational memory and cache

BI & Semantic Layers

Power BI: Executive and visual dashboards (Figma)
Azure AS: Advanced DAX and partitioning
Trino: Distributed semantic layer
Modeling: Star Schema, Snowflake, KPIs

Software & Methodologies

Full-stack: Next.js 16, Node.js (Queues/Workers)
Infrastructure: Docker, Nginx, Linux VPS
Integrations: REST APIs, Evolution API, SFTP
Methodologies: Scrum, Azure DevOps, CI/CD

Supply Chain & Business

SAP Planning (MRP): Raw material management and follow-up
Strategic Procurement: Kraljic Matrix and advanced negotiation
Procurement Operations: Cost savings and efficiency strategies
Business Development: B2B sales and commercial growth

What I Bring to the Business

My guidelines after years on the frontlines of operations and data.

🧠

Focus on the Problem, Not the Tool

Many fall in love with technologies. My focus is on understanding the client's pain and solving it. The client mostly doesn't care about the tech stack; they care that the problem is solved.

💡

Consultative Critical Thinking

Simple "data task execution" is not enough. Analyzing data patterns, understanding the business behind the tables, and not just "accepting" what the client asks for when it doesn't solve the root problem.

💰

Cost-Oriented Architecture

People adopt hyped technologies, convinced by salespeople, without considering the necessary trade-offs. Every architectural decision I make heavily weighs the cost-benefit, keeping the environment lean and highly efficient.

🌐

The Generalist Profile

Understanding everything from the purchase order to the semantic model in Azure, I have the "big picture" view. I know exactly which paths to take and which pieces to connect to deliver the final solution and orchestrate Agentic AI.