Salary: 80K – 150KIndustry: Health-Tech / AILocation: Remote-First (Spain/UK) with a vibrant hub in BarcelonaStack: Python, FastAPI, asyncio, pydantic, RAG, Multi-agent LLMs, GCP, Redis, Postgres, Docker, Kubernetes, OTELThe CompanyYou will be joining a high-growth Health-Technology venture based in Barcelona that has just hit a major inflection point. They are the innovators behind a cutting-edge AI assistant that acts as a digital partner for clinical teams, recently going live with major health systems across Europe.They are scaling rapidly to meet significant healthcare contracts, moving from MVP to large-scale production. This is an international team of clinical experts and elite engineers who believe AI should solve mission-critical medical problems in real-time. If you want to build the technical foundation of a future leader in healthcare AI, this is your moment.The RoleAs an AI Engineer (Backend), you will take end-to-end ownership of Cortex, the core AI "brain" architecture powering the entire platform. This is a backend-first role—your mission is to make the AI systems incredibly reliable, ultra-fast, compliant, and deeply observable in production, rather than inventing new ML models from scratch.In this high-autonomy role, you will collaborate directly with a tight-knit, best-in-class AI team to solve some of the hardest challenges in the industry: real-time voice AI, automated AI audits, patient simulators, and safety-critical escalation systems in a highly regulated medical environment.Your ResponsibilitiesScale & Own "Cortex": Architect, deploy, and maintain robust, modular backend systems using clean architectural principles (SOLID, Clean/Hexagonal Architecture, DDD). Manage service boundaries, reliability targets,



and proactively mitigate failure modes.Orchestrate Multi-Agent Systems: Wire up complex multi-agent orchestration, managing routing between agents, shared state, and clean tool interfaces.Optimize the RAG Pipeline: Engineer high-signal retrieval layers (chunking, hybrid search, re-ranking, caching) and relentlessly prove that grounding holds.Build Platform Infrastructure: Develop automated audit/eval pipelines and patient simulators to stress-test agents at scale before they ever hit production.Master Production Engineering: Write lightning-fast, well-tested Python services utilizing FastAPI, asyncio, and pydantic while optimizing queues, caching, and data stores.Drive Deep Observability: Implement OTEL-first tracing across the agent graph, tracking cost, latency, token visibility, and configuring CI gates to catch regressions before they ship.What You NeedMinimum QualificationsBackend Core: 3–5+ years of software engineering experience with expert-level Python, FastAPI, asyncio, and pydantic. You care deeply about how your code behaves in production.Proven AI Track Record: At least 2+ years of demonstrable experience building, scaling, and taking user-facing AI/LLM applications from MVP (0) to production (1).Architectural Depth: Deep understanding of architectural design patterns (SOLID, Event-Driven, DDD) to manage complex system boundaries.Startup DNA: You thrive in high-intensity,



fast-moving environments and know what wearing several hats actually costs.Autonomy: A self-starter who excels with minimal supervision—moving easily between managing agent patterns, eval-driven development, and production software.Nice to HaveReal-Time & Voice: Experience with WebRTC, LiveKit, SIP, VAD, barge-in, or turn-taking.Advanced LLM Tooling: Programmatic prompt optimization techniques and LLM-as-judge evaluation setups.Cloud & DevOps: Familiarity with GCP (Cloud Run, GKE, Pub/Sub, Vertex AI, Cloud Logging/Trace) and infrastructure tools like Terraform or ArgoCD.Industry Savvy: Prior experience handling healthcare data or working within highly regulated/compliant frameworks.Example Problems You'll TackleStand up the AI audit pipeline so evals run automatically on slices of production traffic, with regression gates wired into CI.Build a patient simulator that lets the team stress-test agents at scale before they ever reach a real medical call.Turn complex EHR (Electronic Health Record) integrations into highly reliable, deterministic tools that the agents can safely call.What's in it for You?Meaningful Impact: You aren't just building a generic chatbot; you are engineering safety-critical systems that actively improve how healthcare is delivered.High-Level Ownership: As an early hire, you will shape the technical foundation of a fast-growing startup and tackle genuinely difficult engineering hurdles.Versátil Setup: Enjoy a European remote-first model, backed by occasional visits to a vibrant, high-energy office hub in Barcelona.Flat Hierarchy: Work closely alongside clinical experts and elite engineering talent in an environment free of corporate bureaucracy.Discipline(s): Information TechnologiesJob type: Senior level - 3 to 5 years

📌 AI Engineer (Backend) (Barcelona)
🏢 Blu Selection
📍 Barcelona