Artificial Intelligence Machine Learning Web DevOps

Real-Time Voice Systems: Design and Architecture in 5 Levels

FORMAT: TalkLEVEL: Advanced LANGUAGE: Spanish

Voice systems have advanced rapidly in recent years, but most implementations still stop at demos: simple combinations of Speech-to-Text, language models, and Text-to-Speech that work in controlled environments but fail when facing real-world conditions. This talk proposes a different approach: understanding voice systems as an architecture that evolves through maturity levels, from basic prototypes to real-time production-ready systems. Through a 5-level framework, we'll walk the full path of a Conversational AI system: from integrating basic components, through orchestration challenges (streaming, latency, turn-taking), to less obvious but critical problems like audio quality, robustness, and user experience, reaching real-time architectures with technologies like LiveKit, and finally exploring where the future is headed with end-to-end systems and multimodal agents. The talk is based on real experience building voice systems in production and focuses on engineering decisions more than specific tools. Attendees will leave with a clear understanding of how to design modern voice systems with Python, what problems to anticipate, and how to structure their own architectures to build world-class conversational experiences.

Speaker

Nicolas Danies

Data Science Manager @ Visa

I'm Data Science Manager at Visa, where I lead artificial intelligence projects for the Andean region focused on turning machine learning and GenAI models into real products with business impact. My work centers on closing the gap between research and production: from designing models to deploying them as scalable systems used by banks and companies across multiple countries. My career has been a fast track through the tech ecosystem in Latin America, passing through companies like Mercado Libre and Rappi, where I worked on high-impact problems like fraud, real-time pricing, and large-scale distributed systems. In these roles I've designed high-performance APIs, production models, and platforms that operate on hundreds of millions of transactions. In parallel, I'm co-founder and COO of an AI startup focused on commercial training through speech-to-speech systems, where I'm building modern architectures integrating voice models, LLMs, and real-time systems. Beyond the professional side, I've always been motivated to build community and accelerate technological development in Colombia. I've been an assistant professor at Universidad de los Andes, taught hundreds of people about machine learning and Python systems, and participated in creating a new Data Science program in the country. My interest in speaking at PyCon Colombia comes from that same motivation: sharing real learnings on how to build AI systems in production, and contributing so more developers in the region can build world-class technology from Latin America.

View speaker

Want to know more?

Join PyCon Colombia newsletter and get a complete overview of our events, speakers and community participation.