How to Build and Scale Voice Agents That Actually Work
Voice agents are quickly moving from demo to deployment, but building one that works reliably in the real world is a very different challenge from simply connecting an LLM to a phone call.
In this session, Josep Marc, Co-founder of Prosper AI, will share a practical perspective on what it takes to design, customize, deploy, and scale voice agents in production. Prosper AI is building and deploying voice agents in the healthcare sector at massive scale, in one of the most operationally demanding environments for AI.
🚨Language: English
About this session
Josep brings years of experience across multiple waves of AI, from earlier machine learning systems to today’s LLM-powered and voice-based agents, and has a unique understanding of what it takes to move from promising technology to systems that deliver real business value.
This will not be a coding-heavy session. Instead, the talk will focus on the strategic, product, and operational side of deploying voice agents successfully: how to define the right use cases, gather requirements from customers, translate business needs into agent behavior, customize workflows, and make sure these systems perform well as they scale.
For anyone thinking seriously about voice AI beyond the hype, this session will offer a grounded and highly relevant view from someone building in one of the most ambitious corners of the market.
Who is Josep Marc Mingot?
Josep Marc Mingot is the co-founder of Prosper AI (https://www.getprosper.ai/). Josep has spent years building, training, and deploying AI systems across different generations of the technology stack. Today, he is focused on the next wave: LLM-powered voice agents, and how to make them work at scale in high-stakes, real-world environments.
What you’ll learn
How to identify the right use cases for voice agents
How to gather customer requirements and translate them into agent design
What “customizing” an agent really means in practice
How to think about deployment, performance, reliability, and scale
What product and business teams need to get right before implementation
How to evaluate whether a voice agent is actually solving the problem it was built for
Who should attend
This talk is especially relevant for:
Product managers
AI engineers
AI and software engineers who want a broader understanding of how voice agents are deployed and managed in real business settings
Innovation, strategy, and AI leads
While the session is more oriented toward product and business audiences, technical attendees will also benefit from understanding the operational and strategic decisions behind successful agent deployment.
Why attend
Most conversations around AI agents focus on what is technically possible. This session with Josep focuses on what actually matters when you want to make these systems useful, reliable, and scalable in the real world.
If you are exploring how AI agents can create value for customers, improve operations, or unlock new products, this is the kind of session that helps bridge the gap between AI excitement and real deployment.