AI (Agentic) Engineer
Mercanis
30+ DAYS
Seniority
Midweight
Model
In-Office
Sector
Salary
Undisclosed
Contract
Full-Time
Design, build, and optimize agentic AI systems that automate complex procurement workflows and power intelligent decision-making across the platform. You will work with cutting-edge technologies in Python, LangGraph, and AWS, integrating and fine-tuning models from leading providers such as OpenAI, Anthropic, and AWS Bedrock.
What you'll do
- Design and build agentic AI systems for procurement automation
- Integrate and fine-tune models from OpenAI, Anthropic, and AWS Bedrock
- Work with LangGraph and related LLM orchestration tools
- Deploy and scale AI systems on AWS infrastructure
- Collaborate with distributed teams in a remote-first environment
What you'll need
- Strong professional experience in Python software development
- Proven experience building AI-driven or agentic systems, preferably with LangGraph or similar frameworks
- Solid understanding of Large Language Models (LLMs) and their integration with external APIs or structured data
- Hands-on experience working with LLM providers such as OpenAI, Anthropic, or AWS Bedrock
- Familiarity with cloud-based deployment and scaling, ideally using AWS services
- Strong problem-solving skills with a focus on maintainable, efficient, and scalable code
Nice to have
- Experience with semantic search technologies (vector databases, embeddings, RAG systems)
- Strong background in measuring, evaluating, and optimizing AI system performance with clear metrics and benchmarking
- Experience implementing observability and monitoring solutions for AI systems
- Knowledge of agent-to-agent (A2A) communication patterns and multi-agent system architectures
What they offer
- Real ownership and autonomy in developing core AI components and system design
- Career growth opportunities in applied AI engineering and architecture
- Flexible remote setup with optional travel to offices in Berlin or Barcelona
- Collaborative and innovative culture that values experimentation and continuous learning

