Senior AI Agent & Workflow Engineer

Position ID: 134249

Responsibilities

- Design Agentic Workflows: Architect end-to-end workflows using LangGraph (or similar), breaking complex problems into composable steps, states, and policies
- Problem Decomposition & Solution Design: Translate high-level business problems into structured agentic solutions, defining workflows, components, and trade-offs across multiple design options
- Build Skills, Tools, and Agents: Implement reusable skills, tools, and agents, and compose them into workflows that are maintainable and testable
- Choose the Right Building Block: Demonstrate strong judgement on when to build a skill vs. a tool vs. an agent vs. a workflow, based on problem decomposition, system design trade-offs, and long-term maintainability
- Model Strategy & Integration: Integrate and operate multiple LLM providers, optimizing for latency, cost, and output quality
- Performance Optimization: Optimize prompt engineering, token usage, chunking strategies, and inference efficiency across the platform
- Evaluations & Quality: Define and implement evaluation strategies (offline and online), datasets, metrics, and guardrails to validate agent behaviour and measure regression over time
- Data Structuring & Validation: Design typed schemas (e.g., Pydantic models) for reliable data flows, tool contracts, and structured outputs
- Observability & Debugging: Implement logging, tracing, and monitoring for agentic systems using tools such as Langfuse or LangSmith
- AI‑Accelerated Engineering: Use AI assistants (including Claude Code) in day‑to‑day work for prototyping, refactoring, testing, and documentation—while applying strong engineering judgement
- Cross‑functional Collaboration: Partner with data scientists, engineers, and product to translate requirements into robust agentic capabilities and deliver measurable value
- Work alongside experts in various technological domains




- Enhance your skills with real‑world experience
- Contribute to a team that values your input and creativity

Qualifications - Must Have

- 5+ years Python development with strong software engineering fundamentals
- Hands‑on experience building production systems with Large Language Models
- High ownership & proactive communication: delivers assigned outcomes autonomously, keeping progress visible and proactively surfacing blockers and trade‑offs
- Proven experience with LangGraph or similar agentic frameworks (LangChain, other orchestration frameworks acceptable)
- Strong understanding of evaluation methodologies for LLM/agent quality (datasets, metrics, human review, regression testing)
- Ability to reason about when to build a skill vs tool vs agent vs workflow, with clear trade‑offs
- Proven ability to decompose complex, ambiguous business problems into scalable technical solutions, from high‑level requirements to detailed design

Qualifications - Nice to Have

- Comfort using AI assistants (e.g., Claude Code) as part of the daily engineering workflow
- Familiarity with LLM observability tools (Langfuse, LangSmith, etc.)
- Background with vector databases and RAG patterns
- Understanding of cost optimisation and token accounting in LLM systems
- Production experience with async Python patterns (asyncio, concurrent request handling)
- Experience with Azure deployment and cloud‑native architectures
- Knowledge of evaluation frameworks and metrics for AI output quality

What You'll Bring

- Problem Decomposition & Solution Design: Ability to take ambiguous,



high‑level business problems and systematically break them down into actionable components, defining multiple solution approaches and selecting the most appropriate one based on trade‑offs
- Clearly defined workflows (agents, tools, skills, orchestration)
- Technical specifications and contracts for each component
- Evaluation criteria, success metrics, and test strategy
- Architectural thinking: Ability to take ill‑defined, complex AI problems and decompose them into clean, modular components
- Production mindset: Comfortable shipping robust, observable, maintainable AI systems—not prototypes
- LLM fluency: Deep practical knowledge of how to work effectively with large language models
- Systems perspective: Understanding of how AI components integrate with APIs, databases, async workers, and monitoring
- Initiative: Comfortable owning full‑stack design decisions and driving technical direction with the team

Benefits

- Competitive salary and yearly bonus
- Versátil working model to support work‑life balance
- Option to work abroad up to 25 days yearly
- Over 300 euros to set up your home office and an additional monthly home office allowance
- Wide range of internal and external trainings, including English, German and Spanish classes
- Ticket restaurant and Health Insurance with flexibility to exchange it for other benefits
- Life and accident insurance
- Collective Life retirement Plan
- 2000 referral bonus if you bring other talented people to the company
- Special banking and insurance conditions plus exclusive employee discounts
- Functional diversity benefits
- Stock options and mortgage benefits

Location

Barcelona, Poblenou

Equal Opportunity Employer

At Zurich, we are an equal opportunity employer. We attract and retain the best‑qualified individuals available, regardless of race/ethnicity, religion, gender, sexual orientation, age, or disability.

📌 Senior AI Agent & Workflow Engineer (Madrid)
🏢 Zurich 56 Company
📍 Madrid