Research Engineer, Frontier Speculative Decoding

📋 Description

• Design and iterate speculator algorithms for accuracy–efficiency tradeoffs.
• Bridge raw data to production models, impacting customer success.
• Work in a fast-paced role at the edge of generative AI.
• Collaborate with experts to solve real-world high-performance challenges.
• Collaborate with customers and ML teams to deploy solutions.
• Own challenging problems with a culture of technical ownership.

🎯 Requirements

• Love for data curation and meticulous attention to detail.
• Experience with hyperparameter searches and tuning trade-offs.
• Experience with existing training codebases; navigate and improve code.
• Strong attention to evaluating checkpoints for quality, performance, reliability.
• Experience with Python and PyTorch.
• Familiarity with SLURM and/or Kubernetes HPC job management.

🎁 Benefits

• Startup equity.
• Health insurance and other benefits.