🛠 Tech Stack

💼 About This Role

You'll join a small team building the safety layer for AI systems, turning petabytes of unstructured text into structured insights. You'll ship pipelines and self-serve analytics that directly influence how AI safety is built. Work deeply on hard problems and see your work ship to production quickly.

🎯 What You'll Do

• Turn petabytes of unstructured text into a structured, explorable view.
• Build scalable representation pipelines for embeddings and search.
• Use LLMs pragmatically for labeling, classification, and data enrichment.
• Deliver insights that translate into product and operational actions.
• Ship self-serve analytics datasets and dashboards.

📋 Requirements

• Strong Python + SQL with an engineering mindset.
• Solid applied NLP/ML experience on real-world text.
• Comfortable with distributed processing and large-scale storage.
• Experience evaluating fuzzy problems with offline/online metrics.

✨ Nice to Have

• Prior work with safety/moderation datasets.
• Experience with policy/rule systems.
• High-volume logging/observability background.

🎁 Benefits & Perks

• 💰 Competitive salary $80K-$150K + equity
• 🏖️ 20 days paid vacation
• 🏢 Hybrid work from Paris + relocation package
• 🏥 Best medical insurance in France
• 🛠️ All hardware, tools, and services you need

📨 Hiring Process

Estimated timeline: 2-3 weeks

1. 1Intro call with colleague· 30 min
2. 2Take-home assignment· 2-3 hours
3. 3Technical interview· 60 min
4. 4Final call with CEO and CTO· 45 min