Surf with... Galtea AI 🏄

Meet the founders of Galtea, the first AI Evaluation platform that ensures AI systems are reliable, secure, and fair

Welcome to a new edition of Surf the wAIve! 🌊 Today, we’re diving deep into the story of a startup that’s leading the way in the AI evaluation space.

The goal of this Substack isn’t just to talk about generic opportunities, but also to uncover hidden gems 💎—those early-stage companies building the foundations of tomorrow’s AI ecosystem.

One of those gems is Galtea AI, a startup developing the essential infrastructure that enables AI to function as expected. And they’re doing it not from Silicon Valley, but from Europe—right from the heart of Barcelona. In the gold rush era, there was a popular saying: “The winners weren’t just the miners, but the ones selling the shovels” 🛠️ The same applies to AI—beyond the LLMs and flashy models, we need the tools that help extract real value. Galtea is building exactly that.

This isn’t PR—it’s a chance to meet real founders, hear their stories, and understand the why behind what they’re building. I met Galtea’s founding team a year ago, and was instantly impressed by their vision of the AI future. Today, I’m proud to support them as an investor through Wayra. So pick your board 🏄‍♂️ and let’s go surf the wAIve with Galtea 🤙

TL;DR Building Trust in Artificial Intelligence

What if you could spot risks before real users ever interact with your GenAI system? Galtea AI empowers enterprises to use AI with confidence. ✅ How? Generating test cases and simulating thousands of synthetic users to evaluate your product, helping you uncover edge cases, vulnerabilities, and gaps before deployment.💪

Founded in October 2024 by Jorge Palomar and Baybars Kulebi, this startup began as a spin-off from the Barcelona Supercomputing Center (BSC) and has quickly emerged from stealth to become a pioneer in the AI evaluation space for startups and enterprises. It has already attracted funding from Tier 1 investors including Wayra, Abac Capital, and JME Ventures and multiple customers across industries are already using their technology.

Galtea AI: Redefining Generative AI Evaluation

In today’s rapidly evolving AI landscape, ensuring the safety, reliability, and compliance of generative AI models is more critical than ever—and that’s exactly Galtea’s mission.

Galtea AI aims to revolutionize how companies develop and deploy generative AI products by offering an end-to-end platform that enables robust evaluation, governance, and compliance throughout the entire lifecycle of AI models.

As generative AI adoption accelerates, so do concerns around issues like model accuracy, security vulnerabilities, bias, and regulatory compliance. The introduction of frameworks like the EU AI Act adds further complexity, requiring organizations to meet stringent standards for high-risk applications or face significant penalties. For businesses leveraging GenAI in sensitive sectors such as finance or healthcare, ensuring their models are trustworthy and compliant is no longer optional—it’s essential.

How it started? 🌱

Galtea is not just another AI tool—it is the result of years of deep AI research and enterprise collaboration. The two co-founders, Jorge Palomar and Baybars Kulebi, worked alongside leading researchers at Barcelona Supercomputing Center (BSC) to develop core AI technologies that now power their platform.

In October 2024, they spun off from BSC to launch Galtea, with the mission of transferring the technology they had built—originally designed to help scientists evaluate AI models—and making it accessible to enterprise developers.

Click PLAY and hear about their story here 👇

How It Works: Galtea’s Solution Overview

“Our technology bridges the gap between experimentation and production,” says Jorge Palomar. Galtea’s innovative platform provides a structured approach to evaluating and governing generative AI systems:

1. Generate Test Cases and Synthetic Agents
   Galtea creates test cases and synthetic agents that simulate real-world user interactions. These simulations are designed to replicate how actual users might engage with your generative AI system, providing insights into its performance under realistic conditions.

2. Evaluate Product Behavior Across Diverse Scenarios
   The platform evaluates your product’s behavior across a wide range of scenarios. This process ensures that the system performs reliably even in edge cases or under challenging conditions.

3. Extract Key Metrics to Identify Risks
   Galtea extracts critical metrics to uncover risks, vulnerabilities, and edge cases. By analyzing these metrics, businesses can identify potential weaknesses in their systems and address them before deployment.

This comprehensive process allows organizations to refine their generative AI models with confidence, ensuring robust performance and compliance with regulatory standards.

“By combining scientific rigor with practical implementation, we set a new standard in AI governance.” says Baybars Kulebi: “Building an AI model is easy—ensuring it works as intended is the real challenge.”

In this video, you’ll hear directly from Baybars, co-founder and CTO of Galtea, as he explains the need for Galtea’s product and how it works.

The goal: Increase safe AI adoption 📈

Galtea’s objective is to accelerate enterprise AI adoption while addressing key challenges such as reliability, risk mitigation, compliance, and time-to-market pressures. The platform productizes advanced AI evaluation technologies into an enterprise-ready framework that helps organizations transition from experimentation to production with confidence.👐

In this video, Jorge breaks down the 4 phases of a Generative AI project and shows how Galtea helps —with tools for standardized testing, benchmarking, monitoring, and feedback loops—to make it easier for teams to adopt and scale AI.

At the edge of new technologies

By moving fast 🚀, bringing together top engineering talent 💡, and continuing to collaborate with BSC on cutting-edge R&D 🧠, Galtea ensures its platform evolves with the latest advancements in AI evaluation—seamlessly integrating new models and capabilities into its SaaS platform, helping clients stay ahead in a rapidly changing AI landscape.

Galtea’s platform tackles critical challenges across multiple industries. Typical use cases include the evaluation of customer-facing chatbots, AI agents, generative AI workflows, document screening, data extraction, and more.

Why It Matters ✨

As generative AI rapidly reshapes the way we work, create, and interact, the question is no longer just what AI can do—but whether we can truly trust it. From customer-facing chatbots to autonomous agents, AI systems are making decisions that impact real people, in real time. And with that power comes responsibility. ⚖️

Galtea empowers enterprises to build that trust. By providing the tools to rigorously evaluate, monitor, and govern generative AI systems, Galtea helps companies ensure that their AI is not only performant, but also safe, fair, and aligned with human values. 💡 In a world where ethical concerns and regulatory pressures are mounting, Galtea offers a practical path toward responsible AI adoption. ✅

Rooted in years of deep research and driven by a mission to scale innovation responsibly, Galtea is redefining how organizations can confidently deploy GenAI—without compromising on reliability, security, or integrity. 🔐 In doing so, it accelerates innovation cycles by up to 80% 🚀 while ensuring that AI remains a tool for progress, not risk.

Because building AI is no longer just a technical challenge. It’s a societal one. 🌍

I hope you enjoyed the post! Don’t forget to subscribe to read more stories about startups riding the next wAIve of innovation.