Member of Technical Staff - Post-Training
Reflection12 months ago
Responsibilities
- Build systems that transform pretrained models into aligned and general agents.
- Drive research and engineering initiatives in post-training, including data curation and large-scale optimization.
- Develop data-generation pipelines, reward models, reinforcement-learning algorithms, and inference-time scaling techniques.
- Collaborate with pre-training and post-training teams to deliver substantial gains in model capability.
- Advance understanding of how large models learn to reason, follow instructions, and improve through reinforcement learning.
Requirements
- Deep understanding of machine-learning fundamentals and practical experience with large-scale LLM training.
- Strong engineering skills and comfort working in complex machine-learning codebases and distributed systems.
- Experience improving model behavior through data, reward modeling, or reinforcement-learning techniques.
- Evidence of owning ambitious research or engineering agendas that produced measurable model improvements.
- Ability to work across research and infrastructure boundaries in a fast-paced startup environment.
- Strong communication and collaboration skills, with a passion for advancing the frontier of intelligence.
Benefits
- Top-tier compensation and equity, with no specific base salary stated.
- Comprehensive medical, dental, vision, life, and disability insurance.
- Fully paid parental leave for all new parents, including adoptive and surrogate journeys, plus family-planning financial support.
- Paid time off and relocation support.
- Daily lunch and dinner, regular off-sites, and team celebrations.
Categories
AI ResearchML Engineering
About Reflection
Reflection is a New York–based, privately held research lab developing open foundational AI models and agentic coding tools for developers, enterprises, and public-sector users. The team includes former researchers from DeepMind, OpenAI, and Anthropic, and their work focuses on transparent, customizable systems that organizations can deploy with ownership and control.