AI / LLM Engineer

About the job

CIBOS Innovation is looking for a hands-on AI / LLM Engineer to design, build, and ship production agentic AI systems — from architecture through deployment to operational tuning. You'll work across CIBOS's commercial products and EU-funded R&D projects, building LLM agents and retrieval systems that solve real business problems for real users, not just demos.

This role suits someone who wants to own AI features end-to-end: design the agent, wire it into live APIs and data stores, ship it to production, and keep tuning it once it's live.

Responsibilities

• Design, build, and deploy LLM-based agents (e.g. LangGraph, ReAct-style tool-calling) that interact with live APIs, databases, and business systems

• Build and maintain RAG and retrieval pipelines — vector embeddings, semantic search, similarity-based deduplication — for tasks like knowledge retrieval, document processing, and inbound-request triage

• Design evaluation frameworks and deterministic scoring for agentic AI behavior, to benchmark and continuously improve system quality

• Own AI features end-to-end: architecture, backend integration (APIs, SQL, async pipelines), deployment, and post-launch tuning

• Integrate AI components into existing product and client systems (REST/SOAP APIs, SQL backing stores, message queues)

• Collaborate with the product and engineering team to scope, prioritize, and ship AI-driven features across CIBOS's product lines and client projects

• Produce clear technical documentation (specs, architecture notes) to communicate AI system design to technical and non-technical stakeholders

• Stay current on emerging LLM/agentic-AI techniques and tooling, and bring relevant ones into CIBOS's stack

Required Skills

• BSc or Master's in Computer Science, Electrical/Computer Engineering, or a related field

• Hands-on experience building and deploying LLM agents or agentic workflows (tool-calling, ReAct-style patterns, frameworks such as LangGraph or equivalent)

• Practical experience with RAG, vector embeddings, and similarity/retrieval systems

• Experience working with LLM APIs (OpenAI or similar) and/or open-source LLMs

• Solid backend engineering skills: Python, API development (e.g. FastAPI), SQL

• Comfortable owning a feature from design through production deployment and iteration

• Experience with Git and standard version control workflows

• Strong analytical and problem-solving skills; ability to work independently and in a small, fast-moving team

• Fluent in Greek and English, with strong technical communication skills

Nice-to-have Skills

• Experience designing evaluation frameworks or benchmarks for agentic AI systems

• Familiarity with Model Context Protocol (MCP) servers for connecting agents to external tools and data sources

• Familiarity with classical ML/forecasting (e.g. LightGBM, scikit-learn, TensorFlow) for hybrid AI + ML product features

• Experience with OCR / document AI and human-in-the-loop review workflows

• Knowledge of distributed/streaming technologies (Kafka, async processing)

• Container experience (Docker; Kubernetes a plus)

• Prior exposure to EU-funded research projects (Horizon Europe / H2020) or working in a research-adjacent environment

• Modern software development practices (Agile/Scrum)

About us

CIBOS Innovation IKE is a deep-tech spin-off of ICCS/NTUA, founded in 2020 and based in Athens, Greece. We work across AI/ML, AR/VR, digital platforms, and edge computing, combining EU-funded research (Horizon Europe, H2020) with real commercial deployments — from AR/voice-guided warehouse picking to LLM-based digital assistants and IoT/digital-twin platforms.

We're a small, senior team that takes products from research prototype to production, and we work closely with major research institutions and industry partners across Europe.

What we offer

• Competitive compensation [range 30-40K gross depending on skills and past experience in related fields]

• Hybrid work arrangement, Athens-based

• Laptop and standard equipment provided

• Opportunity to own AI features end-to-end, from architecture to production

• Exposure to both cutting-edge EU-funded research and real commercial deployments

• Small, senior team — direct ownership and fast decision-making

• A clear path of progression tailored to your skills and interests