Senior Software Engineer - Generative AI & Agent Systems (Remote)

Further Details

Invox Medical is seeking a skilled Software Engineer with expertise in Generative AI and Agent Systems for a 100% remote role based in Catalunya, Spain. This position focuses on designing and implementing production-ready AI agents that can handle clinical workflows with high levels of reliability and security. The ideal candidate will have proven experience in building multi-step agent workflows, managing state, integrating tools securely, and implementing robust observability and evaluation mechanisms. This is a fantastic chance to contribute to a mission-driven company that aims to give healthcare professionals more time for patient care through advanced AI solutions.

Senior Software Engineer - Generative AI & Agent Systems (Remote)

Invox Medical is at the forefront of health technology, developing advanced voice and AI products designed to give healthcare professionals more time for patient care. We are expanding our innovative platform with sophisticated agent systems capable of understanding context, utilizing tools, and executing clinical workflows from start to finish. This is a unique opportunity to join a mission-driven company and contribute to the development of AI solutions that prioritize quality, security, and traceability.

We are seeking a highly skilled Senior Software Engineer with a specialization in Generative AI and Agent Systems. This fully remote position requires a candidate who can translate clinical use cases into reliable, measurable, and maintainable AI agents. You will be responsible for the entire agent runtime, from architecture and state management to production deployment and ongoing operation, applying rigorous software engineering principles to complex, non-deterministic systems.

• Job Title: Senior Software Engineer - Generative AI & Agent Systems
• Location: Remote (Based in Catalunya, Spain)
• Company: Invox Medical
• Key Responsibilities: Designing and building AI agents, managing state and memory, secure tool integration, RAG implementation, model selection, observability, and production deployment.
• Required Skills: Strong Python & TypeScript, LangGraph expertise, production engineering, LLM evaluation, tool calling, RAG application, and a commitment to healthcare standards.
• Career Opportunity: Contribute to cutting-edge AI in healthcare, work with a dedicated team, and help shape the future of medical technology.

Skills

• Experience building production-ready agents with multi-step workflows, cycles, routing, tool usage, persistent state, and explicit failure handling
• Proficiency in LangGraph for designing agent runtimes, including state graphs, nodes, edges, subgraphs, checkpoints, interrupts, and resumption
• Strong asynchronous Python skills, including typing, testing, and service design
• Practical TypeScript experience for platform integrations and APIs
• Experience with production engineering practices: APIs, queues, events, concurrency, idempotency, security, CI/CD, Docker, and AWS deployments
• Familiarity with LLM/agent observability and evaluation tools (e.g., OpenTelemetry, Arize Phoenix, Langfuse, LangSmith, Datadog, Ragas, DeepEval)
• Expertise in designing tool calling schemas, validating model responses, and controlling permissions, errors, retries, and side effects
• Understanding of RAG principles: chunking, embeddings, hybrid search, reranking, security filters, and retrieval evaluation
• Commitment to traceability, privacy, reproducibility, and safety in AI-assisted decisions within the healthcare environment
• Experience with long-term memory, checkpoints, and human-in-the-loop systems for agents

Duties and Responsibilities:

• Design and build production-ready AI agents capable of understanding context, using tools, and executing clinical workflows
• Architect the complete runtime of agents, including state, memory, tool integration, planning, execution, and failure recovery
• Develop agent runtimes using LangGraph, modeling state graphs, nodes, routing, loops, parallelism, streaming, interrupts, and subgraphs
• Model typed and persistent state, session memory, long-term memory, checkpoints, execution resumption, idempotency, timeouts, retries, and cost limits
• Design tool contracts, structured outputs, and tool calling mechanisms, integrating APIs and clinical sources with permissions, validation, sandboxing, and human-in-the-loop intervention
• Build hybrid retrieval, reranking, tenant filters, Graph-RAG, and grounding strategies over medical knowledge, including vector databases and knowledge graphs
• Select and combine models based on quality, latency, cost, privacy, and availability, working with providers like AWS Bedrock and other model APIs
• Instrument each execution for observability of decisions, tools, tokens, cost, latency, and errors, creating datasets and regression evaluations for medical fidelity, groundedness, safety, and tool use quality
• Implement the core agent logic in Python and platform integrations in TypeScript, deploying robust services on AWS and containerized or on-premise environments
• Ensure rigorous software engineering practices are applied to non-deterministic systems

How to apply

• 💼 Prepare your updated Curriculum Vitae (CV) detailing your relevant experience in Generative AI and Agent Systems.
• 🎓 Ensure your qualifications and skills align with the requirements outlined in the job description.
• 🔎 Review the job details carefully to understand the responsibilities and technical expectations.
• 👉 Visit the Eurojobs application portal to submit your application.
• 🌐 Navigate to the provided application link: https://eurojobs.com/jobs/36952138/apply
• 📝 Complete the online application form accurately and thoroughly.
• ⏳ Submit your application before the closing date of December 18, 2026.
• ✅ Successful candidates will be contacted for further assessment.