Vitória, ES · Brazil

Computer Engennier

Henrique Rocha.

Building production-grade ML systems for agritech and industry. Computer Vision · MLOps · Data Engineering.

Hackathon
·
3+Yrs. Exp.
·
1Article
·
2Postgrad.

01 / 05

YOLO v5/v8 object detection and EfficientNet classification trained on agricultural datasets. ONNX Runtime inference for low-latency edge deployment. Research published at IMVIP.

YOLO v8OpenCVEfficientNetONNX
Computer Vision
MLOps & Data Eng.
LLMs & GenAI
Agritech & Satellite
Backend & Systems

Work

IMVIP — Agricultural Risk Platform

AgritechComputer VisionSatellite

AI-powered platform for rural credit risk assessment using EfficientNet leaf disease classification and Sentinel-2 satellite scoring. Delivers an explainable AQR score (0–100) across 7 risk components for banks, cooperatives, and insurers.

Leaf Disease Classifier

Deep LearningONNXPython

Coffee leaf disease classifier trained on BRACOL + RoCoLe datasets with EfficientNet-B2, exported to ONNX Runtime for low-latency edge inference. Research published as scientific article.

Satellite Risk Engine

Remote SensingPythonGIS

Multi-layer risk pipeline combining NDVI, climate indices (ERA5/CHIRPS/SPI), SoilGrids data, and environmental compliance checks into a structured farm-level risk report.

Object Detection System

Computer VisionYOLOGCP

Production object detection pipeline using YOLO v8 and OpenCV for agricultural field monitoring, integrated with GCP and orchestrated via Apache Airflow for high-volume data processing.

KDMile 2026

ML for Agricultural Risk Assessment

Machine learning approach to rural credit risk scoring combining satellite-derived NDVI, climate indices (SPI/CHIRPS/ERA5), SoilGrids data, and computer vision disease detection into a structured farm-level risk report.

Download PDF
ML for Agricultural Risk Assessment
Computer Vision in Agritech Field Monitoring
AQR — Annotation Quality Ranking for Object Detection
IMVIP — Coffee Leaf Disease Platform

What is AQR?

Annotation Quality Ranking

Train on 25% of the data. Keep ≥97% of the mAP.

AQR is a two-stage data selection method for object detection. Instead of training on the full dataset, it ranks images by annotation quality before training — isolating the most informative 25% of samples. The core hypothesis: selecting the right data beats training on everything.

ρ +1.0PlantDoc
·
7/7ρ > 0
·
≥97%mAP @25%

Pipeline

100%
Full Dataset
──▶
AQR-1
Gaussian filter
μ=1.91 · σ=1.75
−34% removed
──▶
~66%
Compatible
──▶
AQR-2
richness × 0.60
precision × 0.40
top Q4 selected
──▶
25%
Q4 · Train
Stage 1 — AQR-1 · Distribution Filter

Removes images whose bounding-box density is incompatible with the test set. A Gaussian score centered on μ_test=1.91 eliminates ~34% of images, keeping only the distributional-compatible subset.

exp(−0.5 × ((n − μ) / σ)²) ≥ 0.50
Stage 2 — AQR-2 · Quality Ranking

Within the compatible subset, images are ranked by richness (bbox count diversity, 60%) + precision (bbox centrality and area quality, 40%). The top quartile (Q4) is selected for training.

0.60 × richness + 0.40 × precision
H1 — Efficiency⚠ conditional

Q4v5 (25% of data) reaches ≥97% of full-dataset mAP. Confirmed: VOC 2012 (97.5%) and CGIAR Wheat (97.9%) with YOLOv8s. Conditional on datasets with ≥2000 images per quartile.

H2 — Generalization✓ confirmed

AQR-v5 ranks training data consistently across domains. Confirmed 7/7 dataset×model combinations. ρ̄ = +0.867. Perfect monotonic correlation ρ = +1.000 on PlantDoc (p < 0.05).

Skills

AI & Computer Vision

YOLO v5/v8OpenCVObject DetectionSegmentationEfficientNetONNX

Machine Learning, DL & LLMs

PythonPyTorchTensorFlowScikit-learnLangChainRAGAI AgentsPandasNumPy

Data Engineering

Apache AirflowDBTSQLOracleMongoDBKafkaETL/ELT

MLOps & Cloud

GCPDockerGitHub ActionsAzure DevOpsFirebaseCI/CD

Backend & Architecture

PythonC#ASP.NET CoreMicroservicesDDDREST APISwagger

Monitoring & Methods

KibanaGrafanaScrumKanbanDesign PatternsDocumentation

Highlights

Hackathon Champion · 4× winner across different competitions
Scientific Article · Identification of Rust on Coffee Leaves Using Computer Vision
B.Sc. Computer Engineering · FAESA · 2025
MBA in Software Architecture · Full Cycle · 2025
Postgraduate in AI · IFES · 2026 (in progress)

About

AI & Data Engineer with hands-on experience in Computer Vision, MLOps, and end-to-end data pipelines in production across agribusiness and industrial sectors. Built an automated object detection system with YOLO/OpenCV on GCP and published research on identifying rust in coffee leaves. Solid engineering base in microservices, DDD, Docker, and CI/CD — combined with expertise in Python, Apache Airflow, DBT, and SQL.

Let's Talk

Open to consulting, research collaboration, and full-time roles.

mailto ↗