Generative AI Engineer

About the company

• EPAM is a leading global provider of digital platform engineering and development services

Responsibilities

• Run supervised fine-tuning and parameter-efficient adaptation (LoRA/QLoRA, adapters) and preference-based alignment (DPO, ORPO, RLHF-style methods)
• Own the dataset lifecycle — sourcing and curating examples, designing annotation and review workflows, deduplication, decontamination
• Generate and filter synthetic data, and version every dataset used in training
• Reduce cost and latency through distillation, quantization, batching and caching strategy, and model routing
• Package and deploy tuned models on Vertex AI (or equivalent), manage model and prompt versioning
• Monitor drift and quality decay in production, define retraining trigger and rollback path
• Implement and tune safety, scope, and PII controls
• Partner with commercial teams to convert qualitative expectations into training signal and measurable eval criteria

Requirements

• 5+ years of hands-on experience with LLM fine-tuning — SFT and PEFT/LoRA at minimum, ideally preference optimization (DPO or similar)
• At least 1 year of relevant leadership experience
• Proficiency in Hugging Face Transformers/PEFT/TRL, Axolotl, or Unsloth, or managed tuning on Vertex AI / Bedrock / OpenAI
• Understanding of training mechanics: tokenization, context length and packing, learning-rate and hyperparameter choices
• Knowledge of catastrophic forgetting, overfitting on small domain sets, GPU/TPU memory constraints
• Expertise in LLM evaluation — designing eval sets and metrics for generative tasks
• Skills in data curation: labeling guidelines, inter-annotator agreement, leakage control
• Familiarity with synthetic data pitfalls and filtering strategies
• Background in LLMOps: model and prompt versioning, drift monitoring, retraining lifecycle management

Conditions

• Remote work (Belarus, Georgia, Armenia, Kazakhstan, Kyrgyzstan, Uzbekistan)