Modern Data Security Engineering: from theory to practice
An intensive in-person workshop on security and privacy applied to data systems, pipelines and platforms.
This training is delivered in Portuguese (pt-BR). All sessions, materials and exercises are in Portuguese.
To present the fundamentals, techniques and decision criteria for handling sensitive data in real systems, connecting concepts from security, privacy and software engineering.
Given the limited number of hours and the breadth of the topics covered, the goal is not to exhaust each technique in depth. The focus is on building repertoire, introducing initial practices, and helping participants assess risk, choose suitable approaches, and continue studying with more clarity after the sessions.
WHO IT IS FOR
- Data engineers
- Software engineers
- Data scientists
- Computer scientists
- Security engineers
- Technology professionals working with data pipelines
SYLLABUS
-
50min
Introduction to the practice
- Information security fundamentals
- Security primitives
- Threat modeling
-
1h00
Methods for sharing identifiers
- Direct and indirect identifiers
- Case study: secure data sharing
- Common vulnerabilities in data matching strategies
-
1h20
Data masking and obfuscation
- Masking patterns and techniques
- Case study: vulnerability demonstration
- Implementing masking policies
-
1h00
Synthetic data
- Synthetic data generation techniques
- Risks: when synthetic data exposes real information
- Case study: failures in test data generation over large datasets
- Tools and frameworks
-
1h10
Differential privacy
- Essential mathematical foundations
- Implementation in data pipelines
- Trade-offs between utility and privacy
- Hands-on demonstration
-
50min
Federated learning
- Core concepts of federated learning
- Architectures and practical applications
- Advantages and limitations
-
40min
Homomorphic encryption
- Concepts and applications
- Computing over encrypted data
- Performance limitations and use cases
-
40min
Integrative hands-on exercise
- Hands-on implementation integrating the workshop concepts
- Discussion of solutions and good practices
METHODOLOGY
- Concepts presented through real industry cases
- Hands-on implementation demonstrations
- Guided discussions on applicability
- Analysis of vulnerabilities and their solutions
- Final practical exercise
MATERIALS PROVIDED
- Slides
- Workbook
- Certificate of completion
PREREQUISITES
- Basic knowledge of data engineering or software engineering
- Familiarity with processing pipelines
- Working knowledge of at least one programming language: Python, PHP, JavaScript, Rust or similar
Content is calibrated to the room during the sessions, going deeper according to the participants' experience and keeping the exercises close to real-world situations.
TECHNICAL REQUIREMENTS
- Your own workstation running Linux, macOS or Windows
- Minimum of 8GB of RAM
- Between 10GB and 20GB of free disk space
- Linux is recommended as the primary environment
INSTRUCTOR
Heitor Gouvêa
Information security researcher with a background in software engineering. For nearly a decade he has worked in contexts involving data protection and application security, covering threat modeling, vulnerability identification, security tooling, and the development of practices for risk reduction.
Over his career he has worked at major Brazilian companies such as Globo and Nubank, as well as with security vendors and in applied research. As a researcher, he has reported vulnerabilities to companies and projects including Activision, MISP, D-Link and Oracle, among others.
LOCATION
Zona Sul, Rio de Janeiro. The exact address is shared only with confirmed participants.
INVESTMENT
This first cohort is offered free of charge, with seats limited to a small group of 6 to 8 participants.
The purpose of this inaugural edition is to validate the content in the classroom, observe how the exercises play out, and make fine adjustments before opening future paid cohorts.