Applied AI Engineer

Maisa (Making AI Accountable) builds an agentic process automation platform that helps enterprises automate complex, decision-heavy processes that traditional automation can't handle and generic GenAI can't be trusted with. Powered by their KPU computing platform, Maisa combines AI-driven problem solving with programmatic execution so every action is reliable, auditable, and enterprise-scale.

About the role: You'll join the Applied AI Research team building production-ready AI agent systems on top of the KPU technology. This is a hands-on builder role focused on implementing, evaluating, optimizing and deploying agentic workflows that solve real enterprise problems, bridging research and production.

What you'll do:

• Design and implement end-to-end AI agent workflows using the KPU technology, working directly with customers to rapidly iterate on solutions.
• Build robust evaluation datasets and frameworks to measure agent/AI performance, reliability, and accuracy.
• Optimize workflows for performance, explainability, and enterprise-grade reliability.
• Push the boundaries of what's possible with LLMs, RAG systems, and AI agents in production environments.
• Stay at the forefront of Agentic AI development, continuously applying new techniques to improve the platform.

What you'll bring:

• Production experience with LLMs and AI features.
• Proficiency in Python for production systems.
• Expertise in prompt optimization and context engineering.
• Understanding of ML evaluation methodologies.
• A self-starter comfortable in ambiguous, fast-moving startup environments.

Fully remote within Europe. Maisa is a Spain-based, Creandum/Forgepoint-backed startup with offices in Barcelona, Madrid and Valencia.