Agentic AI Engineer/AI Agents Backend Developer – Hermes First

We are looking for a highly technical, hands-on, and autonomous Agentic AI Engineer / AI Agents Backend Developer with practical experience in building and orchestrating AI agents in real projects.

The ideal candidate has solid software development/backend experience, good knowledge of Python, and experience with APIs, databases, tool/function calling, and architectures for connecting agents to tools and data, such as MCP or similar solutions.

It is essential to have effectively built end-to-end multi-agent systems, where agents communicate with each other, use real tools and data sources, and execute actions. The candidate must understand state, memory, and context management, as well as concepts like RAG, embeddings, and vector databases.

We are looking for someone capable of building robust systems, using logging, retries, error handling, observability, and guardrails. Experience with Hermes is a major advantage; alternatively, serious experience with other agent orchestration frameworks or architectures and the ability to quickly adapt to a Hermes-first ecosystem is required.

We particularly appreciate experience with multi-agent systems in production, vector databases, Redis, queues, event-driven architectures, Docker, cloud infrastructure, CRM/ERP integrations, and AI evals.

More than the number of years of experience, we are interested in what you have actually built. We are looking for a technical, curious, autonomous, and construction-oriented engineer who can concretely present AI projects and agentic systems developed end-to-end.

We are looking for an Agentic AI Engineer / AI Agents Backend Developer who has already built real AI agents and has experience in orchestrating them in functional systems.

Responsibilities

Creation and development of specialized AI agents for various company functions;
Orchestration of multi-agent systems, where agents communicate with each other, use tools, and execute actions;
Working within a Hermes-first architecture;
Connecting agents to APIs, databases, CRM/ERP, and other internal or external systems;
Building workflows where agents receive information, make decisions, call tools, and execute processes;
Working with memory, state, permissions, retries, logging, observability, and guardrails;
Integrating knowledge sources through RAG, embeddings, and vector databases, when the architecture requires it;
Contributing to the development of an AI-first infrastructure, not just point-in-time integration of LLMs.

About the role

Real experience with the development and orchestration of AI agents is essential. Direct experience with Hermes is a major advantage. If you haven't worked directly with Hermes, it is important to have serious experience with other agent orchestration frameworks or architectures and the ability to quickly enter a Hermes ecosystem.

We are less interested in the number of years of experience and more in what you have actually built. We are looking for a highly technical, autonomous, curious, and construction-oriented person.

What we offer:
- Permanent individual employment contract;
- Market-level salary + meal tickets worth 40 RON/working day + bonuses;
- Working hours Mon-Fri: 8:30 AM - 5:30 PM, with a 1-hour lunch break included;
- We provide transport from Carrefour Pantelimon and the Republica metro station.