Hybrid AI Engineer

at AdTech Holding

• Job title: AI Engineer
• Job type: Full Time
• Experience level: Mid level, Senior
• Salary: Not specified
• Degree requirement: No Education Requirement
• Tech skills: Cloud, Python
• Location requirements: Hybrid, Limassol, Cyprus

About the role

• AI Engineer designing and developing effective LLM workflows for AdTech projects in Cyprus or Serbia. Collaborating with ML engineers on data preparation and fine-tuning tasks.

Responsibilities

• Design, develop, and iteratively improve effective prompts, chains, and agentic workflows for a variety of tasks (e.g., data categorization, content generation, code assistance)
• Work with open-source LLMs (e.g., Llama, Mistral) to adapt them for specific internal tasks using prompt-based techniques and parameter tuning
• Collaborate with ML engineers on the initial data preparation and evaluation for potential fine-tuning projects
• Create robust and user-friendly interfaces for internal AI tools, such as chatbots or text processors
• Working with intelligent features — smart prompts in tasks and autonomous agents that gather backlogs and predict risks
• Complex production pipelines — designing and scaling end-to-end systems from data to inference, ensuring seamless operation and rapid feature delivery
• Designing and implementing systems based on Retrieval-Augmented Generation (RAG)

Requirements

• Strong experience in prompt engineering and designing effective LLM workflows (e.g., using Chain-of-Thought, few-shot learning)
• Solid Python programming skills for scripting, API interactions, and data manipulation
• Practical, hands-on experience with the OpenAI API and/or other commercial LLM APIs
• A solid conceptual understanding of how Large Language Models work (transformers, attention, tokenization, limitations)
• Familiarity with frameworks like LangChain or LlamaIndex to build complex applications (Nice-to-Have)
• Basic experience with running inference of open-source LLMs locally or on cloud instances (Nice-to-Have)
• Understanding of Retrieval-Augmented Generation (RAG) architecture and its implementation (Nice-to-Have)
• Experience with the Hugging Face ecosystem (transformers library, loading models, running inference) (Nice-to-Have)

Benefits

• health insurance
• fully covered 10 working days of sick leave per year
• 5 additional day-offs per year
• English and other languages compensation
• quartile and annual bonuses
• corporate events
• gifts
• food (yummy!) - you can find snacks in our worldwide working spaces and breakfasts & lunches in the offices!