XConf Americas 2025

by technologists, for technologists

More details coming soon.

As AI transforms how teams build software, the challenge is to innovate without losing control or compromising quality. This talk explores how to apply AI-first practices while keeping developers at the center of the process. Based on the RIPER-5 model, a Spec-Driven Design approach, real-world examples will be shared that combine AI innovation with solid delivery principles. Attendees will gain practical insights to navigate between hype and reality, maintain human oversight, and integrate AI responsibly, ensuring that reliability and innovation advance side by side.

The majority of AI initiatives fail because the data supply chain is fragmented, slow, and missing machine-readable governance, leaving data assets without the context and control needed to produce high-quality, AI-ready data products. AI Building for AI addresses this constraint by using Zap, a system powered by specialized LLM agents that automate the data product lifecycle and reduce delivery time from weeks to hours, as demonstrated with a life sciences client. This approach shifts data product teams from traditional builders to strategic reviewers focused on value and governance, supported by an agent-driven architecture and robust evaluation frameworks that enable unprecedented speed, trust and scalability for enterprise AI.

The LATAM Airlines journey reflects the evolution from an agility-driven model focused on speed and delivery to a product operations model centered on long-term strategic value. As the market matured, the organization realized that being efficient in launching features wasn’t enough to drive meaningful business results. This talk explores how Product Operations became the catalyst for that transformation, going beyond theory to establish a product organization greater than the sum of its parts—one capable of delivering measurable impact for both customers and the business.

At a leading Canadian fintech, an AI-first Software Engineering Transformation program is redefining how teams build and deliver software. The initiative introduces a new framework and set of practices that integrate AI across the software development lifecycle while addressing the cultural and organizational challenges that come with large-scale change. The session explores the program’s rationale, phased rollout, and measurable impact on productivity and engineering effectiveness.

This panel brings together different perspectives on the inclusion of people with disabilities, featuring Carolina Ignarra, founder of Talento Incluir, along with leaders from the Globo account team and an executive from Globo itself. Together, they explore how inclusion strategies move from intention to action, showing how accessibility and representation drive innovation within organizations. The discussion highlights real examples of collaboration between consulting teams, internal leadership, and clients to create work environments where technology truly serves everyone.

We live in a world increasingly mediated by algorithms—from how we consume content to decisions about credit, work, and opportunities. This talk invites reflection on how artificial intelligence can be used ethically, responsibly, and inclusively, exploring its real impact on our lives, the dilemmas of automating human decisions, and global frameworks that aim to ensure transparency and fairness. Through practical examples and case studies, it demonstrates that it’s possible to develop technology that respects diversity and promotes social justice.

Caixa sought to modernize its credit risk model pipelines, and during project analysis, an opportunity emerged that went beyond data: designing a Data Mesh-based infrastructure architecture. After aligning with cloud and security teams, criteria and approaches were defined to move forward with the project, conceived as an MVP with the potential to significantly accelerate model implementation. Through collaboration across teams, a structure was created to maximize benefits and open new growth opportunities. The lessons learned provide valuable insights into applying MLOps and Data Mesh practices in complex government contexts.

Driving innovation in a highly regulated, high-responsibility cost center is no small challenge. This talk explains how Thoughtworks partnered with LATAM Airlines’ Maintenance division to create an Experimentation Lab designed to generate measurable business value. The session highlights a framework that brings together Data, Design, and AI to build stakeholder confidence, accelerate delivery, and enable data-driven decision making, resulting in strong business impact and laying the foundation for intelligent digital transformation.

More details coming soon.

The vision of an autonomous cloud—where infrastructure provisions, scales, protects, and optimizes itself—is moving from concept to reality as AI-driven operations advance. This session explores how AI is transforming the foundations of GitOps, observability, and automation into a new generation of adaptive, self-managing platforms. It will feature examples of how AIOps, machine learning, and reinforcement learning are being used to detect anomalies, predict failures, and optimize workloads in real time. The session also distinguishes what’s achievable today from what remains experimental, offering a practical roadmap toward truly intelligent infrastructure.

Creating and maintaining technical documentation is often a slow, manual process that few people enjoy. This talk demonstrates how artificial intelligence can turn that effort into an intelligent, efficient, and consistent workflow. Using the MDX-GDocs case, it showcases how an agent built with Gemini and Model Context Protocol (MCP) converts text or Word files into MDX documentation with Docusaurus components, ready for automatic repository integration via Pull Request. Beyond the technical demo, it shares lessons on how to apply AI ethically and effectively in engineering teams, reducing friction, improving the developer experience, and freeing time for what truly matters: building.

More details coming soon.