About Us
Holoai Ltd. is an AI startup founded in 2024, building with Joaia at the frontier of travel and discovery. We give tourism destinations across Europe the tools to unlock their local knowledge and turn it into agentic AI guides that reason, remember, and adapt. Destinations build and own their guides themselves. Travelers get something that didn't exist before.
Imagine a guide that knows every hidden bar in Zurich's alternative scene, a Roman merchant who brings the ancient origins of Chur to life, or a carnival insider who walks you through Basel's Fasnacht tradition with the depth of someone who has lived it for decades. Each guide has its own personality, its own knowledge domain, its own way of telling a story. They can be fictional characters, digital twins of historical figures, local experts, or thematic specialists. The possibilities are just beginning to open up.
We work with destinations like Zurich Tourism, Basel Tourism, Lucerne Tourism, Graubünden Ferien, and more are joining every month. We are a small, focused team on a mission to make the world's local knowledge accessible to every traveler, so that anyone, anywhere, can discover a place with the depth and curiosity it deserves. If that mission excites you, come build it with us.
The Role
You'll be the first dedicated AI Engineer at Joaia, working directly with Roger (CTO & co-founder). You own the Guide Engine - the AI core that powers every guide, story, and recommendation our users experience.
The Guide Engine is already in production, powering real guide experiences for destinations across Europe. It runs on a network of specialized agents, each doing a specific job, working together to deliver something that feels intelligent, local, and alive. We've come a long way. And there's a huge amount still to figure out.
This is a role for someone who is deeply curious about how these systems behave, who has strong intuitions about where they break, and who wants to push the boundaries of what's possible. Better reasoning, sharper outputs, stronger evaluation frameworks, agentic capabilities, memory - the foundation is solid and the next level is where it gets interesting.
What You'll Do
-
Own the AI logic behind the Guide Engine and drive its continuous improvement
-
Develop, test, and refine prompts across different guide workflows
-
Build automated LLM benchmarking and evaluation systems
-
Evaluate and compare models for quality, consistency, and fit
-
Measure the impact of prompt, model, and workflow changes on output quality
-
Improve how the system uses context, structured knowledge, and local data
-
Define how guides reason across place, story, event, and recommendation
-
Strengthen guardrails and identify vulnerabilities in AI outputs
-
Work closely with Product and Engineering to turn AI concepts into scalable features
What We're Looking For
-
Proven hands-on experience with LLMs, prompting, and AI workflows - professionally or through substantial personal projects
-
Deep understanding of how LLMs behave: strengths, failure modes, inconsistencies, and how to steer them
-
Strong logical and analytical thinking - you break complex problems into clear systems
-
A feel for language, nuance, and meaning, and how wording changes model behavior
-
Understanding of conversation design, dialogue logic, and user intent
-
Self-driven and genuinely curious - you've explored and built in this space on your own terms
-
No specific degree required - we care about what you've actually done
What We Offer
-
Real ownership and direct impact on the product from day one
-
Equity / stock options - you're building this alongside us, not just working in it
-
A small, honest team where decisions get made fast and your voice matters
-
Hybrid work setup based in Zurich
-
Work at the frontier of AI - agentic systems, memory, real-world deployment, and a lot still to discover
How to Apply
Send us a short message via email to darius@joaia.app and tell us why you fit. Attach your CV and optional a link to relevant work.
We take hiring seriously. Our process includes structured evaluation with real tasks, because this role requires demonstrated judgment, not just a strong interview.
And one more thing: we know the perfect candidate on paper doesn't always exist. If you don't tick every box but genuinely believe you could make a difference here, apply anyway. We care more about how you think and what you've built than whether your CV matches every line of this description.
