I build systems that take AI research into production. My work sits at the intersection of two tracks I have followed in parallel since early on:

• Applied AI research: computer vision and GenAI, peer reviewed in IEEE and Springer (stroke detection from wearable signals at 99% accuracy, medical image analysis, license plate recognition)
• Scalable backend engineering: microservices and distributed systems in production, currently at Zup (Itaú Group)

I care less about a headline accuracy number and more about what it takes to make a model survive real data: noise, latency, cost, and monitoring.

Backend and architecture: Java/Spring · Go · Node · Python · microservices · gRPC/GraphQL · Kafka · AWS AI/ML: Computer Vision · Deep Learning · GenAI · LLMs/RAG · MLOps (MLflow, FastAPI) Data and infra: Databricks · Docker · Terraform · ELK

Roadmap of focused, Dockerized projects, each with a real README and a benchmark. Pinned as they ship.

• 🔬 yolo-training-pipeline: computer vision training and inference benchmark, Dockerized
• 🤖 llm-eval-harness: reproducible evaluation suite for RAG and LLM pipelines
• ☁️ mini-aws-emulator: lightweight local emulator for core AWS services (Go)
• 🏗️ spring-hexagonal-payments: hexagonal architecture reference implementation

If you work on applied AI or on the systems that put it in production, I am always glad to exchange ideas.

I build systems that take AI research into production. My work sits at the intersection of two tracks I have followed in parallel since early on:

• Applied AI research: computer vision and GenAI, peer reviewed in IEEE and Springer (stroke detection from wearable signals at 99% accuracy, medical image analysis, license plate recognition)
• Scalable backend engineering: microservices and distributed systems in production, currently at Zup (Itaú Group)

I care less about a headline accuracy number and more about what it takes to make a model survive real data: noise, latency, cost, and monitoring.

Backend and architecture: Java/Spring · Go · Node · Python · microservices · gRPC/GraphQL · Kafka · AWS AI/ML: Computer Vision · Deep Learning · GenAI · LLMs/RAG · MLOps (MLflow, FastAPI) Data and infra: Databricks · Docker · Terraform · ELK

Roadmap of focused, Dockerized projects, each with a real README and a benchmark. Pinned as they ship.

• 🔬 yolo-training-pipeline: computer vision training and inference benchmark, Dockerized
• 🤖 llm-eval-harness: reproducible evaluation suite for RAG and LLM pipelines
• ☁️ mini-aws-emulator: lightweight local emulator for core AWS services (Go)
• 🏗️ spring-hexagonal-payments: hexagonal architecture reference implementation

If you work on applied AI or on the systems that put it in production, I am always glad to exchange ideas.