AI Engineer

Juryo; Barcelona; On-site; EUR 90,000 - 130,000; Full time

Benefits offered by this job

Joining a rocket ship

Small, demanding team with great vibe

Freedom to do things your way

Unlimited tokens to build with tools

Job summary

Juryo is building an AI voice operating system for B2C law firms. We are looking for AI Engineers to own how our agents think, talk and decide, with real production calls and no demos.

You’ll design agent architecture and reduce latency, delivering natural, robust conversations in Spanish and English. You will stay on top of models, build reusable components, and ensure we ship reliable tools across voice and WhatsApp.

Qualifications

• You’ve built and shipped LLM-based products in production.
• Know how models fail: hallucinations, prompt injection, broken tool calls and how to fix.
• Able to code in TypeScript or Python and work with APIs/webhooks.
• AI native: you use coding agents daily and know when not to trust them.
• Fluent in English; Spanish is a big plus.

Responsibilities

• Design and improve agent architecture: prompting, tool calling, memory, multi-agent flows across voice and WhatsApp.
• Build evals and testing to measure agent performance before client notice.
• Push latency down and naturalness up in voice, including turn taking and speech recognition in Spanish and English.
• Turn learnings from Forward Deployed Engineers into reusable components for faster onboarding of new firms.
• Stay on top of models and providers and decide what goes into production.

Skills

LLM product shipping

Model failure handling

Coding with TS/Python

AI native developer

English fluency

Barcelona living

Tools

TypeScript

Python

APIs

Webhooks

Job description

At Juryo we are building the AI voice operating system for B2C law firms. Our agents qualify, sell, onboard, support clients and handle collections. They already work with more than 30 law firms in Spain and the UK, and in 7 months we have gone from 0 to €1M ARR.

The role

Our agents talk to thousands of people every week: someone who just had an accident, someone drowning in debt, someone trying to get their papers in order. Every conversation either moves a case forward or loses it. We are looking for AI Engineers to make our agents the best in the market at that.

What is an AI Engineer at Juryo? You own how our agents think, talk and decide. Not demos, real production calls with real clients, where latency, tone and a single wrong tool call make the difference between a signed client and a hang up. We build AI native: you’ll direct coding agents to ship faster and spend your time where your judgement matters.

If you’ve shipped LLM products that real users depend on, or you’re obsessed with getting there, this is for you.

What you’ll do

• Design and improve our agent architecture: prompting, tool calling, memory, multi agent flows across voice and WhatsApp.
• Build the evals and testing that tell us if an agent is getting better or worse, before a client notices.
• Push latency down and naturalness up in voice: turn taking, interruptions, voices, speech recognition in Spanish and English.
• Turn what our Forward Deployed Engineers learn on the ground into reusable agent components, so every new firm starts from something better.
• Stay on top of models and providers, test what’s new and decide what goes into production.

What we’re looking for

• You’ve built and shipped LLM based products in production, not just prototypes.
• You know how models fail: hallucinations, prompt injection, broken tool calls. And you know how to measure and fix it.
• You’re comfortable writing code: TypeScript or Python, APIs, webhooks.
• You’re AI native. You use coding agents daily and know when not to trust them.
• You’re hungry to work. We’re a team that wants to make its mark, and we put in the hours.
• You speak English fluently. Spanish is a big plus, since most of our agents speak it.
• You live in Barcelona or are willing to move here. We work from the office, 100% on site.
• Nice to have: multimodel AI experience

What we work with

TypeScript, Python, ElevenLabs, Retell, OpenAI, Anthropic, PostgreSQL, webhooks, Vercel. If you know some of them, great. If not, they can be learned: we care more about how fast you learn than about what your CV says.

What we offer

• Joining a rocket ship as part of the founding team, with still a lot left to build
• A small, demanding team with a great vibe, where you’ll learn in one year what would take you five anywhere else.
• Freedom to do things your way, without layers of approval.
• Practically unlimited tokens to build with the best tools out there.