About the role
EverAI is hiring an AI LLM Engineer to improve open-weight models powering its AI companionship platform, which serves more than 50 million users. The role focuses on post-training and model behavior, with work spanning data, evaluation, experimentation, and production deployment. The engineer will collaborate with a small AI team and colleagues in product, web, and DevOps.
Responsibilities include fine-tuning with SFT and LoRA/QLoRA, applying methods such as DPO and GRPO, diagnosing model failures, building training data and evaluations, and using real user behavior to guide iterations. Applicants should have personally post-trained open-weight LLMs for a user-facing product and shipped them to production, alongside hands-on experience with Python, PyTorch, Hugging Face, and GPU training. The listing describes this as fully remote in Europe and prefers B2B contracts while remaining flexible.
Highlights
- Fine-tune and post-train open-weight LLMs for a large consumer product
- Build training data and evaluations that influence model development
- Work on personalization, consistency, multilingual behavior, and group interactions
- Remote in Europe, with four weeks of paid time off
This recap is dataskew's editorial summary, not the company's copy.