We are looking for an AI Engineer to join our client's team. In this role, you will have a chance to grow your technical and soft skills, and build a thorough expertise of the industry of our client.

This role is available only in Bucharest and involves an employment contract (CIM), with hybrid work model (2 days per week on-site).

Responsibilities:

• Architect, build, and operate complex multi‑agent systems using frameworks like Google ADK and LangGraph.
• Design LLM‑powered agents, including orchestration logic, tool use, memory, and human‑in‑the‑loop flows.
• Implement reliable state‑machine or graph‑based orchestration with retries, fallbacks, and escalation paths.
• Develop and maintain RAG pipelines, tool‑calling workflows, and effective prompting strategies.
• Evaluate and improve agent performance using ML principles, safety checks, and curated datasets.
• Integrate agents with surrounding infrastructure (vector databases, queues, observability tools, LLM libraries).
• Deploy, monitor, and troubleshoot AI systems in production, including CI/CD, logging, and tracing.
• Diagnose and resolve complex issues across the AI stack (models, prompts, APIs, tools, data).
• Collaborate with product teams to translate requirements into agent capabilities.
• Produce clear documentation for architectures, decisions, and operational runbooks.

Requirements:

• 3+ years in software/ML/AI engineering, including 1–2 years building production LLM systems.
• Degree in Computer Science, Engineering, Mathematics, or equivalent experience.
• Strong Python skills and solid software engineering practices (Git, testing, CI/CD).
• Experience with REST/gRPC APIs.
• Good understanding of ML fundamentals and at least one ML/DL framework.
• Strong understanding of LLMs: prompting, sampling, fine‑tuning, RAG, safety.
• Familiarity with agentic frameworks (ADK, LangGraph) and ability to work with their codebases.
• Experience with vector databases, SQL, and cloud platforms (GCP/AWS/Azure).

Nice to have:

• Experience with hosted LLM platforms, LangChain/LlamaIndex, observability stacks.
• Background in IR, NLP, AI security, or contributions to open-source AI.

To get a better idea of the project you could be a part of, start a discussion with our EDUROM consultant—just click "Easy Apply."

Your application will go to EDUROM Recruitment, part of the EDUROM Group. We acknowledge that moving on to a new job or challenge has a great impact on your professional and personal life. We believe you can make a great career choice based on a well-informed decision. Through honest discussions and relevant interviews, our HR consultants promise to assist you during this important time.