Uma trilha é um evento presencial em São Paulo e tem a duração de um dia inteiro com sete palestras e um painel de discussão.
Trilhas Premium e Stadium têm transmissão ao vivo e as palestras são gravadas. Os demais conteúdos (Expo, Workshops, Mentorias) são somente presenciais.
Thursday, September 18, 2025
09h às 19h
ProMagno
Avenida Professora Ida Kolb - 513 /
Jardim das Laranjeiras
São Paulo -
SP
IN-PERSON OR REMOTE ACCESS WITH ONLINE BROADCAST
Faça sua inscrição em qualquer trilha Premium presencial e ganhe um workshop Bônus e uma mentoria bônus no mesmo dia! Mas não perca tempo pois as vagas são limitadas.
Confira a lista completa de workshops.
Confira a lista completa de mentorias.
Você também ganha acesso Premium à TDC Community por um ano, com acesso a workshops e mentorias semanais.
Inscrição online
1 track:
R$
305 for R$
186
Inscrição online
1 track:
R$
305 for R$
275
Inscrição online
1 track: R$ 305
Inscrição híbrida (presencial + online)
1 track:
R$
490 for R$
310
2 tracks:
R$
840 for R$
560
3 tracks:
R$
1.180 for R$
790
Inscrição híbrida (presencial + online)
1 track:
R$
490 for R$
420
2 tracks:
R$
840 for R$
730
3 tracks:
R$
1.180 for R$
1.070
Inscrição híbrida (presencial + online)
1 track: R$ 490
2
tracks: R$ 840
3
tracks: R$ 1.180
| Programação Stadium: |
|---|
|
08:00 to 08:55 Credenciamento |
|
09:00 to 10:00 Opening of the event and mini keynotes |
| Programação desta Trilha: |
|---|
Track opening by coordination10:10 to 10:25 Here the coordinators introduce themselves and make an introduction to the track. |
Local Development in the AI Era
Kevin Dubois
(Sr Principal Developer Advocate -
IBM)
18/Sep
10:30-11:05
Local Development in the AI EraTrilha AI & EMERGING TECH 18/09/2025 10:30 -11:05ProMagno
Most of us like to do local development. It means we're in control of any dependencies, network issues/latency, configurations and cost. And we love to say that 'It works on my machine' don't we? Now, with the advent of AI-driven development, it has become more challenging than ever before to do pure local development. In this session we'll take a look at various solutions to continue to do local, network-optional development, even with AI:
Kevin Dubois
IBM |
Developer productivity for GenAI apps
Eddú Meléndez
(Software Engineer -
Docker)
18/Sep
11:15-11:50
Developer productivity for GenAI appsTrilha AI & EMERGING TECH 18/09/2025 11:15 -11:50ProMagno
The GenAI landscape has grown a ton in the last year with many technologies and approaches that application developers must master to include AI in their end-user and enterprise apps. However, more focus must be put into enabling standard software development workflows: testing, ensuring RAG correctness and efficiency, local development environment and CI setups, and security of LLM artifacts. In this session, we look at how application developers can integration test their GenAI apps with local models, augmented with RAG both locally and in CI. You'll learn how the setups for GenAI apps in the inner development loop can be programmatically managed. 
Eddú Meléndez
Docker |
Dapr - Solving the challenges of Distributed Systems
Matheus Cruz
(Sr Software Engineer -
ITAÚ)
Mauricio Salatino
(Principal Software Developer -
Diagrid)
18/Sep
11:55-12:30
Dapr - Solving the challenges of Distributed SystemsTrilha AI & EMERGING TECH 18/09/2025 11:55 -12:30ProMagno
Building distributed systems doesn't have to be complex. In this session, you'll learn how Dapr (Distributed Application Runtime) can streamline microservices development by abstracting common concerns like service invocation, state management, pub/sub, and security. We'll explore how Dapr integrates with the Cloud-Native ecosystem and how it helps you focus on business logic instead of boilerplate infrastructure code. 
Matheus Cruz
ITAÚ 
Mauricio Salatino
Diagrid |
|
Intervalo para almoço 18/Sep
12:30-13:50
|
The Future of AI is Event-Driven: Building Reactive and Scalable Multi-Agent Systems
Daniel Takabayashi
(Engineering Manager -
CONFLUENT)
18/Sep
13:50-14:25
The Future of AI is Event-Driven: Building Reactive and Scalable Multi-Agent SystemsTrilha AI & EMERGING TECH 18/09/2025 13:50 -14:25ProMagno
With the rise of LLMs, AI Agents have become a reality. The next challenge lies in orchestrating complex multi-agent systems. But is the traditional Request-Response paradigm up to the task? In this talk, we'll explore this model's limitations and present Event-Driven Architecture (EDA) as the key to building truly reactive, scalable, and resilient AI ecosystems. Learn how to leverage events to decouple your agents and create future-proof autonomous systems. This is a session focused on software architecture for the age of AI. 
Daniel Takabayashi
CONFLUENT |
Shaping the Next Wave: Facing AIs Challenges and Becoming Builders of the Future
Elder Moraes
(Principal Developer Advocate -
IBM)
Kevin Dubois
(Sr Principal Developer Advocate -
IBM)
Stephen Chin
( -
Neo4j)
Dra. Anna Medeiros
( -
Airia)
Víctor Orozco
(CTO -
Nabenik)
18/Sep
14:30-15:30
Shaping the Next Wave: Facing AIs Challenges and Becoming Builders of the FutureTrilha AI & EMERGING TECH 18/09/2025 14:30 -15:30ProMagno
AI is transforming how we build systems, but it also raises tough challenges for developers and organizations. This panel brings together leading industry experts to share different perspectives on issues like bias in models, regulation, infrastructure demands, and the new skills engineers need to adapt. Instead of a single viewpoint, the audience will hear contrasting experiences and approaches from practitioners who are directly involved in shaping the future of AI. |
|
Networking e Visitação a Stands 18/Sep
15:35-15:50
|
Architectures for Building AI Agents
Fabiane Nardon
(Diretora de Inteligência de Dados -
TOTVS)
18/Sep
15:55-16:30
Architectures for Building AI AgentsTrilha AI & EMERGING TECH 18/09/2025 15:55 -16:30ProMagno The rise of AI agents is transforming how we design and build software. In this talk, well explore architectural patterns that combine deterministic and non-deterministic approaches, changing how developers reason about systems and decision-making. Well cover strategies for building AI agents in a cost-efficient way, including managing token usage and optimizing resource consumption. A key focus will be on how you provide data to the agent, an often overlooked but critical piece for both performance and reliability. The session also dives into practical techniques for creating Model Context Protocol (MCP) servers more efficiently, as well as essential security considerations such as protecting the data layer to prevent prompt injection. By the end, youll walk away with architectural principles, cost-saving practices, and security patterns that make AI agents practical, safe, and ready for real-world deployment. 
Fabiane Nardon
TOTVS |
Supercharge agentic AI apps: A DevEx approach to cloud-native scaffolding
Daniel Oh
(Red Hat -
IBM)
18/Sep
16:35-17:10
Supercharge agentic AI apps: A DevEx approach to cloud-native scaffoldingTrilha AI & EMERGING TECH 18/09/2025 16:35 -17:10ProMagno
This session will demonstrate live how this framework speeds up the development of agentic artificial intelligence (AI) applications. Key features include automated code generation, pre-configured infrastructure, and integration with popular AI libraries. These tools help developers reduce set-up time and focus on building intelligent functionality. By the end of this session, you will understand how cloud-native scaffolding streamlines agentic AI development. You?ll gain practical knowledge that will help you use these tools to build intelligent agents faster and more effectively. 
Daniel Oh
IBM |
|
Applying the OWASP Top 10 to Agentic System Security
Dra. Anna Medeiros
( -
Airia)
18/Sep
17:15-17:50
Applying the OWASP Top 10 to Agentic System SecurityTrilha AI & EMERGING TECH 18/09/2025 17:15 -17:50ProMagno Agentic AI can automate complex workflows, but without proper safeguards, these agents risk exposing credentials, leaking sensitive data, or behave unpredictably. In this talk, I will show you how to apply the OWASP Top 10 for LLMs to agentic systems. What you will learn: 1. How to map agentic-specific threats (e.g., prompt injection, system-prompt leakage, unbounded resource consumption). 2. Key technical controls: input/output validation, least-privilege credential management, and human-in-the-loop checkpoints. 3. Best practices for adversarial testing and real-time logging to detect anomalous agent behavior. 
Dra. Anna Medeiros
Airia |
|
Agentic GraphRAG: AIs Logical Edge
Stephen Chin
( -
Neo4j)
18/Sep
17:55-18:30
Agentic GraphRAG: AIs Logical EdgeTrilha AI & EMERGING TECH 18/09/2025 17:55 -18:30ProMagno
AI models are getting tasked to do increasingly complex and industry specific tasks where different retrieval approaches provide distinct advantages in accuracy, explainability, and cost to execute. GraphRAG retrieval models have become a powerful tool to solve domain specific problems where answers require logical reasoning and correlation that can be aided by graph relationships and proximity algorithms. We will demonstrate how an agent architecture combining RAG and GraphRAG retrieval patterns can bridge the gap in data analysis, strategic planning, and retrieval to solve complex domain specific problems. 
Stephen Chin
Neo4j |
| Programação no final do dia: |
|---|
|
18:30 to 18:35 Encerramento da trilhaOs coordenadores fazem um breve encerramento com agradecimentos. |
