What You'll Do
Lead the technical direction of a generative AI system processing over 80 million tokens daily. Design and implement scalable solutions for context management, memory retrieval, and low-latency inference to power a responsive, real-time conversational experience. Take full ownership of the model lifecycle—from data pipeline design to deployment and monitoring.
Guide strategic decisions on when to use prompting, fine-tuning, or retrieval-augmented generation. Develop and maintain custom classifiers to detect harmful content while preserving user experience in an explicit context. Build intelligent moderation systems that go beyond simple filtering to understand nuance and intent.
Write production-level Python code using PyTorch and modern LLM frameworks, directly contributing to a system used by millions. Define alignment strategies that shape the behavior and personality of AI interactions, balancing technical precision with user engagement.
Requirements
You have 8+ years of engineering experience, with a strong focus on delivering machine learning features at scale. You’ve worked extensively with Python and PyTorch, and are fluent in tools like vLLM, HuggingFace, and evaluation frameworks for LLMs. You’re experienced in shipping production models and iterating quickly based on performance data.
You’re comfortable working with NSFW content and understand the challenges of moderating such environments without degrading the user experience. You have a practical mindset—valuing shippable, effective solutions over theoretical perfection. You act with ownership, tracking metrics, catching regressions, and advocating for the end user long after deployment.
You bring intuition to model alignment: knowing how temperature, sampling, and prompt design influence behavior. You thrive in fast-moving, autonomous environments and are driven by impact, not process.
Benefits
- Fully remote work with flexibility to choose your ideal environment
- 20 days of paid time off per year
- Annual in-person gathering for team connection and strategy
- Monthly health insurance stipend of 100 USD
- Unlimited mental health and lifestyle coaching for you and up to two family members
- Co-working access budget: up to 35 EUR twice per month
- Learning fund for courses, books, conferences, and certifications
- Company-issued laptop and monitor setup allowance up to 250 USD
- Premium access to AI development tools including ChatGPT, Claude Code, Cursor, and Hugging Face
