AI Agent Engineer, Python, LLMs, RAG

This is a fully remote position, open to applicants in Spain.

📋 Description

• Design and develop AI agent architectures

• Collaborate on planning, tool calling, function calling, memory, and context engineering

• Create solutions based on semantic routing, workflows, and agents

• Work with MCP and, preferably, A2A

• Design and develop multi-agent orchestration architectures

• Design and maintain agent runtime, including execution loops, state management, orchestration, retries, and checkpointing

• Implement streaming, human-in-the-loop, observability, tracing, and context management

• Design systems for tool execution and concurrency management

• Create and implement evaluation strategies for agents and LLMs

• Develop offline evaluations, online evaluations, benchmarks, and evaluation datasets

• Implement golden datasets, synthetic datasets, and LLM-as-a-judge

• Develop regression tests and track experiments for reproducibility

• Design and develop RAG solutions, including chunking, embeddings, reranking, and vector databases

• Optimize retrieval quality and the use of context windows

• Engage in LLM Engineering: prompting, structured outputs, JSON Schema, and function calling

• Optimize tokenization, costs, latency, and caching

• Develop backend solutions using Python

• Work with FastAPI, Docker, Kubernetes, Redis, Kafka, and PostgreSQL

• Implement observability solutions: tracing, metrics, logs, and agent traces

• Monitor costs and latency of AI-based systems

• Develop strategies for unit testing, integration, evaluation, and regression

• Integrate AI solutions into CI/CD processes

• Work with cloud environments and specialized AI services

⛳️ Requirements

• At least 3 years of experience in developing Artificial Intelligence solutions

• Experience in agent architectures and multi-agent systems

• Knowledge of planning, tool calling, function calling, memory, context engineering, and semantic routing

• Familiarity with workflows versus agents and MCP (Model Context Protocol)

• Experience in agent runtime: execution loops, state management, orchestration, retries, and checkpointing

• Knowledge of streaming, human-in-the-loop, observability, tracing, and context management

• Experience in evaluations: offline/online evaluations, benchmark design, and regression testing

• Knowledge of golden datasets, synthetic datasets, and LLM-as-a-judge

• Experience with RAG: chunking, embeddings, reranking, and vector databases

• Familiarity with LLM Engineering, prompting, structured outputs, JSON Schema, and function calling

• Knowledge of tokenization, costs, latency, caching, and context windows

• Proficiency in Python is essential

• Experience in backend engineering

• Knowledge of unit testing, integration, evaluation, and regression

• Understanding of observability and monitoring of AI-based applications

• Preferred experience with LangGraph, OpenAI Agents SDK, FastAPI, Redis, Kafka, PostgreSQL, OpenTelemetry, Azure, AWS, or GCP

• Preferred knowledge of A2A, Docker, Kubernetes, MLflow, Grafana, Prometheus, Azure OpenAI, Vertex AI, Amazon Bedrock, AgentCore, and CI/CD pipeline integration

🏝️ Benefits

• 23 vacation days per year

• 14 payments, 12 monthly and two extras (June and December)

• Flexible compensation: childcare vouchers

• Flexible compensation: health and dental insurance

• Flexible working hours

• Work modality can be on-site, remote, or hybrid depending on the situation

• Access to cutting-edge technologies to stay updated