Senior AI LLM Engineer—Fine-Tune Open-Weight LLMs (Remote)

EverAI; United States; Remote; USD 84,000 - 140,000; Full time

Benefits offered by this job

PTO 4 weeks

Annual gathering

Health & wellness stipend

Co-working space budget

Learning budget

Company laptop

Job summary

EverAI is assembled as a fully remote, global AI team to push consumer-grade AI toward safer, more engaging experiences. You will fine-tune open-weight LLMs, guide post-training, and collaborate with data, product, and devops to ship improvements that users feel in conversations.

You should have hands-on experience post-training LLMs for production, a strong Python/PyTorch background, and comfort working with uncensored models for moderation.

Qualifications

• Personally post-trained open-weight LLMs for a user-facing product.
• AI must be core to the product, not just an experimental feature.
• Fine-tuning is central to your experience.
• Strong Python and PyTorch with hands-on GPU training and debugging.
• Build data and evals yourself and bring them into production.

Responsibilities

• Fine-tune open-weight LLMs using SFT, LoRA/QLoRA and post-training methods such as DPO and GRPO.
• Diagnose model failures and turn them into experiments.
• Build training data and evals that shape future models.
• Improve model behavior across personalization, multilingual capabilities, and group interactions.
• Ship to production and use real user behavior to guide iterations.
• Speed up the training loop for faster experiments and reproducibility.

Skills

Post-training LLMs

AI in product

Fine-tuning

Python PyTorch HF

Data & evals

Production deployment

AI tooling

NSFW handling

Tools

PyTorch

Hugging Face

Job description

EverAI is assembled as a fully remote, global AI team to push consumer-grade AI toward safer, more engaging experiences. You will fine-tune open-weight LLMs, guide post-training, and collaborate with data, product, and devops to ship improvements that users feel in conversations.

You should have hands-on experience post-training LLMs for production, a strong Python/PyTorch background, and comfort working with uncensored models for moderation.