About EverAI

EverAI is building the future of AI companionship, serving over 50 million users. Our flagship product is the world’s largest AI companionship platform, governed by our proprietary moderation system, EverGuard—an internal AI designed to ensure everything we build is safe, ethical, and human-first.

Our founding team brings strong entrepreneurial experience building and scaling web products from early stage to IPO and multi-million dollar exits.

The Role

Most LLM roles stop at experimentation. At EverAI, the models you improve power experiences for 50M+ users. Most of our traffic runs on self-hosted, fine-tuned open-weight models. Your job is to take models from failure analysis to production across data, post-training, evals, and model behavior.

You will work closely with our LLM Tech Lead, CTO, Product, Web, and DevOps teams to solve core problems including personalization, character consistency, long conversations, multilingual behavior, and training cycle acceleration.

Key Responsibilities

Model Fine-Tuning: Fine-tune open-weight LLMs using SFT, LoRA/QLoRA, and post-training methods such as DPO and GRPO.
Failure Diagnosis: Diagnose model failures and turn them into experiments across data, post-training, prompting, or classifiers.
Data & Evals: Build training datasets and evaluation frameworks that directly guide subsequent training iterations.
Behavioral Improvement: Enhance model performance across personalization, character consistency, tone, multilingual support, and group interactions.
Production Deployment: Ship models to production and iterate based on real user behavior metrics.
Training Optimization: Accelerate training loops to make experiments faster, reproducible, and easier to compare.

Qualifications

Hands-on experience post-training open-weight LLMs for consumer-facing products and shipping them to production.
Background working in environments where AI is core to the product/business, not just a side experiment.
Extensive experience with model fine-tuning.
Strong proficiency in Python, PyTorch, and Hugging Face, including hands-on GPU training and debugging.
Proven ability to build evaluation pipelines and datasets independently to drive evidence-based decisions.
Capability to transition models from notebooks to high-scale production systems.
Active daily use of modern AI engineering tools.
Comfort working with uncensored models and explicit NSFW/moderation content during safety and evaluation workflows.

Bonus Points

Experience with conversational AI, roleplay applications, or personalized consumer AI products.
Work with multilingual or multimodal LLMs.
Background in AI moderation, safety, or adversarial testing.
Open-source LLM or post-training contributions.

What We Offer

Flexible Contract Options: B2B preferred, open to flexible arrangements for long-term commitment.
100% Remote Work: Work from anywhere that suits your productivity.
Paid Time Off: 4 weeks (20 working days) of PTO annually.
Annual Team Retreat: Yearly in-person company gathering.
Health & Wellness: $100 USD monthly allowance for health insurance, plus unlimited 1:1 sessions via OpenUp for you and up to three family members.
Co-Working Allowance: Budget for up to 2 co-working passes per month ($35 EUR / $40 USD per visit).
Learning & Growth Budget: Dedicated budget for courses, books, conferences, and certifications.
Hardware & Tools: Provided laptop, $250 USD monitor budget, $150 USD peripheral budget, and premium access to ChatGPT, Hugging Face, Claude Code, and other required tools.