AI Engineer

Description

Summary: nuAura is seeking an AI Engineer to build, improve, and evaluate agentic intelligence for their health companion product, focusing on LLM agents and conversational experiences. Highlights: 1. Shape AI agent behavior for personalized health guidance and user support. 2. Improve grounding, reliability, and safety in health-sensitive conversations. 3. Collaborate across teams to integrate AI capabilities into production systems. **About nuAura** nuAura is building a modern health companion that helps people better understand their body through biomarker data, doctor expertise, and personalised daily guidance. Our product brings together advanced health testing, medical interpretation, and practical recommendations across longevity, nutrition, movement, recovery, and mental performance. **Location**: Portugal\-based, Lisbon preferred **Working model**: Remote or hybrid **Employment type**: Full\-time **Contract type**: Permanent or contractor, depending on candidate location and setup **Language**: English required; Portuguese is a plus **Salary**: Competitive, depending on experience **Start date**: As soon as possible **About the Role** We are looking for an AI Engineer to help build, improve, measure, and evaluate the agentic intelligence behind the nuAura product. This is a hands\-on engineering role focused on LLM agents, health\-focused conversational experiences, tool use, memory, evaluation, safety, data quality, and production reliability. You will work closely with backend, product, and clinical stakeholders to make the agent more useful, grounded, trustworthy, measurable, and maintainable. You should be comfortable working with ambiguous product problems, designing robust AI system behaviour, using data to understand quality and failure modes, and turning experimentation into production\-quality software. **What You Will Do** * Build and improve LLM\-powered agent workflows for health guidance and personalised user support. * Design agent behaviour around planning, tool use, memory, retrieval, follow\-up handling, and response generation. * Improve grounding, reliability, safety, and escalation behaviour in health\-sensitive conversations. * Develop evaluation workflows to measure quality, regressions, routing, safety, and multi\-turn conversation behaviour. * Build observability into agent workflows so the team can inspect decisions, tool calls, guardrail outcomes, latency, cost, and failure modes. * Work with structured user data, product signals, and knowledge sources in a responsible way. * Collaborate with backend engineers to integrate agent capabilities into production systems. * Review AI system outputs, diagnose failure modes, and turn qualitative and quantitative findings into clear product and engineering improvements. * Write clean, well\-tested Python code for agent orchestration, model integrations, and supporting infrastructure. * Help define practical metrics, datasets, and feedback loops for improving agent quality over time. * Help document patterns, tradeoffs, and technical decisions so other engineers can understand and contribute. * Keep a pragmatic balance between experimentation, product value, safety, cost, latency, and maintainability. **Must Have** * 4\+ years of professional experience building software systems, preferably with Python. * Hands\-on experience building or integrating AI\-powered product features, LLM workflows, or conversational systems. * Strong engineering fundamentals: clean architecture, testing, debugging, and maintainable code. * Ability to reason about agent behaviour across multi\-step and multi\-turn interactions. * Experience designing or running evaluations for quality, safety, reliability, or regression detection. * Comfortable working with APIs, structured data, databases, and backend service boundaries. * Ability to build observability into software systems so behaviour, failures, and performance can be understood. * Good judgment around AI safety, hallucination risk, grounding, user trust, and when to escalate instead of answer. * Ability to explain complex AI system behaviour clearly to engineers and non engineers. * Product\-minded approach with a focus on shipping useful, reliable user experiences. **Nice to Have** * Background in data science, machine learning engineering, applied statistics, or quantitative product analysis. * Experience with statistical analysis, experiment design, model evaluation, feature quality, or production ML workflows. * Experience turning messy product or user data into useful signals for ranking, recommendations, personalisation, or quality measurement. * Familiarity with model monitoring, drift detection, dataset curation, or feedback loop design. * Experience with FastAPI, Pydantic, async Python, or similar backend tooling. * Experience with OpenAI, Anthropic, Gemini, Azure OpenAI, or other LLM providers. * Experience with agent orchestration, tool registries, memory systems, guardrails, or semantic verification. * Experience with eval harnesses, benchmark design, golden datasets, or regression testing for AI systems. * Experience with AI/agent observability, tracing, evaluation dashboards, production monitoring, or debugging complex model\-driven systems. * Experience working with health, wellness, clinical, biomarker, or other sensitive user data. * Familiarity with cloud environments, CI/CD, containers, and production monitoring. **What Success Looks Like** You are successful in this role when you can improve the quality, trustworthiness, and reliability of the nuAura agent while making its behaviour observable, measurable, understandable, and maintainable for the wider engineering team. This role is for someone who wants to build practical AI systems that users can trust, not just prototypes that work in demos. **Interested?** Send your CV to recruitment@nuaurahealth.com — we’d love to hear from you. Tipo de oferta: Integral/Full\-time Localização do trabalho: Presencial