Hugo Souto's Portfolio

Intro

Hi, I'm Hugo Souto, an AI Solution Engineer and Coordinator at the Ministry of Management and Innovation (MGI), Secretariat of Digital Government, Federal Government of Brazil. I'm a researcher-practitioner with over 7 years of experience in the data science lifecycle, specializing in designing complex machine learning, deep learning, and generative AI architectures.

I lead the development of Brazil's national AI Platform as part of the Brazilian Artificial Intelligence Plan (PBIA), building systems that will serve over 170 million citizens. My work combines systems thinking with hands-on technical implementation, bridging organizational understanding with delivery capability at government scale.

With a solid foundation in management, innovation, and organizational learning (MSc in Learning Organizations + MBA in Innovation), I've proven ability to lead complex projects and present technical solutions to top management. I act as a consultant in AI solution prospecting for various Federal Government agencies and interface with startups, technology companies, and research institutions for project execution.

Previously, I worked as a Senior Data Scientist and Product Owner at ComprasGov, where I created the ComprasGov Insights platform—an ML-based analytics system analyzing decisions on a platform that processed over US$120 billion in purchases over five years. I'm also a former long-distance triathlete, committed to bringing the same focus and discipline to AI development.

Check out some of my current work and technical expertise.

Work

Current Projects

Open Source & Technical Projects

Skills

Expertise

• 🚀 AI Solution Engineer & Coordinator at Ministry of Management and Innovation, leading development of Brazil's national AI Platform serving 200+ million citizens
• 🤖 7+ years in data science lifecycle, specializing in ML, Deep Learning, and Generative AI architectures with systemic vision integrating business, client, and user needs
• 🏛️ Product Owner and lead developer of multiple government-scale AI projects, including ComprasGov Insights (analyzing US$120B+ in procurement)
• 🌎 International AI consultant for government projects (Costa Rica - CLAD partnership) and World Bank Government Analytics Fellowship nominee
• 💰 Coordinator of 90M BRL (~$18M USD) CPQD research contract for AI Factory developmentm, among other projects
• 🎓 MSc in Learning Organizations at UFPB with Data Science research + MBA in Innovation at UFCG/PaqTcPB
• 🎯 Proven ability to lead complex projects and present technical solutions to top management (directorate, secretariat, ministerial, and presidential levels)

Specializations

• Azure AI Engineer Associate Certification (AI-102) - In progress
• Mathematics for Machine Learning and Data Science Specialization by DeepLearning.ai
• Microsoft Azure Data Scientist Associate
• Create Machine Learning Models in Microsoft Azure
• Microsoft Azure Machine Learning for Data Scientists
• Machine Learning Specialization - Stanford Online (2020)
• High Impact Presentations - Dale Carnegie (2025)

Languages, Databases & Clouds

• Python (Advanced - ML/DL/LLMs)
• SQL (Expert - Analytics & Data Engineering)
• PostgreSQL
• Microsoft SQL Server
• Microsoft Azure AI & ML
• Databricks (Azure & Multi-cloud)
• Google Cloud Platform (GCP)

Data Science

• Jupyter Notebook
• Pandas
• Numpy
• Matplotlib
• Seaborn

Machine Learning Frameworks

• Scikit-Learn
• TensorFlow
• PyTorch
• Keras

Generative AI & LLMs

• Large Language Models (LLMs) - GPT, Claude, Gemini
• RAG (Retrieval-Augmented Generation) architectures
• Multi-agent systems & orchestration
• Prompt engineering & optimization
• AI security & guardrails implementation
• LangChain, LlamaIndex, and custom frameworks

Data Engineering, Prototyping & ML Engineering

• Airflow
• Docker
• Streamlit

Data Vizualization & Analytics

• Google Looker
• Qlik Sense
• Power BI
• Tableau

Tools

• VS Code
• GitHub

Posts

Maximizing Real-World Impact with Machine Learning Models: Two Simple Strategies for Success with AutoML

Topics: Data Science, Machine Learning, AutoML

This article explores the critical transition from learning and experimenting with machine learning models in notebooks to applying these skills in real-world business scenarios. It addresses the challenges of working with real-world data and emphasizes the importance of models that can adapt and learn continually. The piece presents practical strategies for bridging the gap between ML theory and real-world impact using AutoML approaches.

Read on Medium | View LinkedIn post

More Content

Please, check my posts at my Medium Blog.

About

I'm an AI Solution Engineer and Coordinator at the Ministry of Management and Innovation, leading the development of Brazil's national AI Platform that serves over 200 million citizens. As a researcher-practitioner, I combine systems thinking with hands-on technical implementation in machine learning, deep learning, and generative AI.

My work focuses on architecting scalable, secure AI infrastructure for multi-agent chatbot systems, RAG implementations, and government-scale AI solutions. I coordinate a significant portion of a 390M BRL (~$78M USD) research contract for the AI Factory initiative and act as a consultant for AI projects across various Federal Government agencies.

With a Master's degree in Learning Organizations and an MBA in Innovation, I bring over 7 years of data science experience to complex challenges at the intersection of technology, business, and public policy. I've successfully presented and defended technical projects to top management at directorate, secretariat, ministerial, and presidential levels.

I'm also a former long-distance triathlete, committed to bringing the same focus and discipline to AI development. Check out my current work and technical expertise.

Contact

• Twitter
• Facebook
• Instagram
• GitHub

Elements

i = 0;

while (!deck.isInOrder()) {
    print 'Iteration ' + i;
    deck.shuffle();
    i++;
}

print 'It took ' + i + ' iterations to sort the deck.';