LLM ENGINEER (CONVERSATIONAL AI) at Pheon
REMOTE (CET) / CYPRUS
RESPONSIBILITIES:
- Collection and preparation of training datasets, fine-tuning and adaptation of LLMs for role-playing dialogues
- Designing systems for long-term memory, context management, and character states
- Building evaluation pipelines to assess character consistency, memory retention, narrative coherence, and user engagement
- Optimizing LLM inference for latency, throughput, and cost metrics
REQUIREMENTS:
- Practical experience with LLM fine-tuning (SFT, LoRA, QLoRA, DPO, RLHF)
- Experience in designing and implementing RAG systems
- Development of LLM evaluation pipelines, benchmarks, and quality metrics
- Work with Context / Memory (long-term memory, dialogue state, personalization)
- Inference optimization (vLLM, TensorRT-LLM, batching, latency, throughput, GPU cost)
CONDITIONS:
- Fixed salary in a stable currency
- Employment under a Service Agreement in the jurisdiction of Cyprus (not RF, not RB)
- Full remote work (CET time zone)
- Complete autonomy and the opportunity to influence company products
- Strong international team and direct impact on business growth and scaling
