AI Researcher — Distillation

Featherless AI is seeking an AI Researcher specialized in model distillation to join our team. You will focus on transforming large, resource-intensive models into efficient, high-performance systems that maintain superior quality, bridging the gap between academic research and real-world production.

Key responsibilities

• Design and evaluate advanced model distillation techniques, including teacher-student training and layer-wise distillation.
• Research and optimize the critical tradeoffs between model size, latency, memory usage, and overall accuracy.
• Develop novel distillation approaches specifically for large language models and inference-constrained environments.
• Conduct large-scale experiments and rigorous ablation studies to validate research outcomes.
• Collaborate with engineering teams to productionize research and publish findings in top-tier venues like NeurIPS, ICML, or ICLR.

Requirements

• Strong background in machine learning research with hands-on experience in model distillation, compression, or quantization.
• Proven publication record in conference, journal, or workshop papers.
• Solid understanding of deep learning fundamentals, including optimization and training dynamics.
• Fluency in PyTorch and experience conducting research-grade experimentation.
• Excellent communication skills with the ability to articulate complex research ideas and limitations.

What we offer

• Significant ownership over research direction within a growing Series A startup environment.
• Strong institutional support for publishing research and contributing to open-source projects.
• Access to substantial compute resources and challenging, production-scale problems.
• Collaboration within a small, highly technical team of experts in machine learning and systems.