Artificial Intelligence Engineer

This role operates in the adult entertainment and NSFW AI space. You will work directly with uncensored models, explicit dialogue, and edge-case safety/moderation filtering., * Fine-tune open-weight LLMs using SFT, LoRA/QLoRA, and post-training alignment (DPO, GRPO).

• Diagnose model failure modes (tone drift, persona breaks, repetition) and resolve them via data curation, alignment, or guardrail classifiers.
• Enhance multi-turn consistency, long-context roleplay, tone, and multilingual behaviour.
• Build automated roleplay evaluation benchmarks and synthetic data engines.
• Deploy models directly to production and optimise training cycles on GPU clusters.

• Proven track record post-training and shipping open-weight LLMs into high-traffic consumer production.
• Deep hands-on experience with Python, PyTorch, Hugging Face, and GPU training/debugging.
• Strong expertise in post-training techniques: SFT, LoRA/QLoRA, and preference tuning (DPO/GRPO).
• Ability to design custom eval frameworks and high-signal training datasets.
• Complete comfort working with uncensored models and explicit NSFW content.

Bonus Points

• Experience in conversational roleplay, character AI, or virtual companions.
• Multilingual or multimodal model training.
• Background in content moderation classifiers, red-teaming, or open-source LLM