Agentic AI Engineer
Mondia Group2 months ago
Madrid, SpainMid Level
Responsibilities
- Implement agent logic, tools, and orchestration against the defined architecture.
- Build and tune RAG pipelines covering ingestion, embeddings, retrieval, re-ranking, and evaluation.
- Design and optimize Postgres schemas, queries, and vector indexes for production workloads.
- Build the test-generation and test-execution engine for the AI Code Pipeline.
- Integrate systems with Git, CI/CD, and internal services.
- Build evaluation harnesses, observability, and guardrails for measurable and safe agents.
- Own the reliability, latency, and cost of shipped systems.
- Write clean, typed, well-tested Python and participate in code review.
Requirements
- At least 4 years of professional software engineering experience with Python, including async programming, typing, packaging, performance, and maintainable code.
- At least 1 year of experience building LLM or agentic systems involving tool use, function calling, orchestration, prompting, and evaluation.
- Strong problem-solving and debugging skills for ambiguous production problems.
- Solid Postgres experience, including schema design, query optimization, and pgvector extensions.
- Experience shipping and operating production services, including testing, CI/CD, observability, and on-call practices.
- Comfort with Git-based workflows and code review.
- Experience automating unit, integration, functional, security, and performance testing.
- Experience integrating with developer tooling and CI/CD systems.
- At least 6 months of demonstrated experience building with Claude Code in a development workflow.
Benefits
- Hybrid Madrid office arrangement with 13 work-from-home days per quarter.
- Company bonus.
- Ticket Restaurant.
- Private health insurance.
- Flat hierarchies and short decision-making paths.
- Collaboration with an international team across three continents.
- Creative, modern office location.
Tech Stack
Categories