I design and build production-grade software systems across data engineering, APIs, distributed architectures, machine learning, and MLOps.

I'm currently working as a Tech Lead, combining hands-on software development with technical leadership, architecture decisions, engineering standards, and delivery of production systems.

My engineering background is strongly rooted in data and machine learning, but my work goes beyond models and pipelines — I build the software infrastructure that makes complex systems reliable, scalable, and maintainable.

I don't just build software. I design the systems that make software work.

┌─────────────────────────────────────────────────────────────┐
│                         TECH LEAD                           │
│                                                             │
│  Architecture · Technical Decisions · Engineering Quality   │
│  Mentorship · Delivery · System Design · Problem Solving    │
└──────────────────────────────┬──────────────────────────────┘
                               │
              ┌────────────────┼────────────────┐
              ▼                ▼                ▼
       ┌──────────────┐ ┌──────────────┐ ┌──────────────┐
       │   Software   │ │     Data     │ │      ML      │
       │  Engineering │ │ Engineering  │ │   & MLOps    │
       └──────┬───────┘ └──────┬───────┘ └──────┬───────┘
              │                │                │
              └────────────────┼────────────────┘
                               ▼
                    Production Systems

I work across the entire engineering lifecycle:

• System architecture and technical design
• Backend and API development
• Data pipelines and distributed processing
• Machine learning systems
• MLOps and model lifecycle
• Database architecture
• Infrastructure and containerization
• Technical leadership and engineering standards

Python · XGBoost · Scikit-learn · SHAP · FastF1 · Pandas

A motorsport analytics platform combining data engineering, machine learning, statistical analysis, and race simulation to model Formula 1 performance.

• End-to-end data ingestion and transformation pipelines
• Custom feature engineering for temporal and motorsport data
• Modular ML architecture
• Model evaluation and experiment comparison
• Explainability with SHAP
• Race simulation and scenario modeling
• Data-driven strategy analysis

• Qualifying grid prediction with ~3 position MAE
• XGBoost selected through comparative model evaluation
• Driver and team performance modeling
• Tire degradation and race strategy modeling
• Continuous model validation and recalibration

🔗 GitHub

Python · FastAPI · PostgreSQL · aiohttp · Playwright · Docker

An open-source data platform for FIA World Endurance Championship data.

Built to provide a reusable engineering layer for endurance racing analytics, similar in spirit to the ecosystem around FastF1.

WEC Data Sources
       │
       ▼
┌─────────────────┐
│ Data Discovery  │
│   & Cataloging  │
└────────┬────────┘
         │
         ▼
┌─────────────────┐
│ Async Ingestion │
│     aiohttp     │
└────────┬────────┘
         │
         ▼
┌─────────────────┐
│ Data Processing │
│   & Validation  │
└────────┬────────┘
         │
         ▼
┌─────────────────┐
│   PostgreSQL    │
└────────┬────────┘
         │
         ▼
┌─────────────────┐
│    FastAPI      │
│       API       │
└─────────────────┘

• Automated historical session discovery
• Asynchronous high-performance data collection
• Reverse engineering of modern web data sources
• Structured season, event, session, classification, and stint datasets
• PostgreSQL data architecture
• REST API for programmatic access
• Designed as an extensible open-source data platform

🔗 openwec.com 🔗 GitHub

Apache Kafka · Spark Structured Streaming · PySpark · MLflow · Docker Compose

An end-to-end distributed streaming architecture for real-time fraud detection.

Event Producer
      │
      ▼
    Kafka
      │
      ▼
Spark Structured Streaming
      │
      ├── Feature Engineering
      │
      └── ML Inference
              │
              ▼
           MLflow

• Real-time event ingestion with Kafka
• Distributed processing using Spark Structured Streaming
• Feature engineering inside the streaming pipeline
• ML inference on streaming data
• Fraud detection optimized for 96% recall
• ROC-AUC improved from 0.53 → 0.77
• Experiment and model tracking with MLflow
• Fully reproducible Docker Compose environment
• Domain-oriented service architecture

🔗 GitHub

FastAPI · OpenAI Whisper · MongoDB · Docker Compose

A distributed audio-processing platform that transforms Formula 1 team radio into searchable structured data.

• Microservices architecture
• FastAPI service layer
• Asynchronous audio processing
• OpenAI Whisper transcription
• MongoDB indexing and full-text search
• Thousands of audio files processed
• ~2.2s average processing time per audio file
• <50ms API response latency
• Containerized development environment
• Modular and maintainable codebase

🔗 GitHub

I believe good engineering is not about choosing the most sophisticated technology.

It's about choosing the right architecture for the problem.

My approach focuses on:

• Simplicity before unnecessary complexity
• Clear system boundaries
• Observable and testable services
• Reliable data flows
• Reproducible environments
• Automation over manual processes
• Designing for maintainability
• Measuring systems instead of guessing

Architecture is not about drawing boxes. It's about making the right trade-offs.

• Backend architecture
• Distributed systems
• API design
• System reliability
• Software architecture

• Real-time data processing
• Machine Learning systems
• MLOps
• Feature engineering
• Model evaluation and explainability

• Technical architecture
• Engineering standards
• Technical decision making
• Mentorship
• Delivery and execution
• Building maintainable engineering teams

• Race simulation
• Telemetry analysis
• Strategy modeling
• Driver and team performance
• Motorsport data infrastructure

I use side projects as engineering laboratories.

Instead of building isolated demos, I try to turn ideas into complete systems with real architecture, APIs, data pipelines, infrastructure, documentation, and reproducible environments.

Some of the areas I explore:

Software Engineering
        │
        ├── Distributed Systems
        ├── APIs
        ├── Data Platforms
        ├── Streaming
        ├── Machine Learning
        ├── MLOps
        └── Simulation
                │
                ▼
        Motorsport Analytics

• 💼 LinkedIn
• 📧 palomacordeiro2009@hotmail.com
• 🔬 Architecture & Experimental Projects

Engineering is the craft. Data is the material. Software is the product.