🛠 Tech Stack

💼 About This Role

You'll join Reflection's data team to ensure high-quality data for training open-weight AI models. You'll operationalize quality standards for reasoning and alignment data, partnering with post-training teams. This role offers the chance to shape data-driven AI breakthroughs at a cutting-edge startup.

🎯 What You'll Do

• Analyze and operationalize quality standards for LLM post-training data
• Translate research requirements into measurable quality signals
• Design LLM-as-a-Judge frameworks for automated data quality checks
• Build reusable QA pipelines to deliver high-quality training data

📋 Requirements

• 3-5 years building data pipelines, QA systems, or evaluation workflows
• Proficiency in Python and building scalable ML/LLM workflows
• Experience with large datasets and automated evaluation systems
• Familiarity with LLM training and evaluation processes

✨ Nice to Have

• Experience with LLM-as-a-Judge frameworks
• Background in post-training or agentic environments
• Experience working with external data vendors

🎁 Benefits & Perks

• 💰 Top-tier salary and equity structured to retain global talent
• 🏥 Comprehensive medical, dental, vision, life, disability insurance
• 👶 Fully paid parental leave and family planning support
• 🏖️ Paid time off and relocation support
• 🍽️ Daily lunch and dinner provided