Senior Machine Learning Engineer (Safety, LLM)

TL;DR

Senior Machine Learning Engineer (Safety, LLM): Building and iterating on next-generation safety systems for Reddit with an accent on developing, training, and optimizing large language models. Focus on scaling enforcement of platform rules, implementing distributed training pipelines, and ensuring responsible AI deployment.

Location: Must be based in the United States (Remote-friendly, with optional access to offices in San Francisco, Los Angeles, New York City, and Chicago).

Salary: $216,700 – $303,400 USD

Company

Reddit is a massive community-driven platform connecting millions of users through shared interests and authentic conversations.

What you will do

• Design, develop, and deploy ML models, specifically LLMs, for NLP tasks.
• Implement and maintain training pipelines using distributed training frameworks.
• Collaborate with cross-functional teams to define model architectures and gather requirements.
• Conduct model evaluations and performance analysis to improve accuracy and reduce bias.
• Stay current with research in NLP, multimodal signals, and large language models.
• Contribute to ethical AI principles and best practices for responsible development.

Requirements

• 5+ years of relevant MLE experience in NLP, deep learning, and AI model development.
• Strong proficiency in Python and deep learning frameworks like TensorFlow, PyTorch, or Hugging Face Transformers.
• Expertise in distributed training frameworks (e.g., Ray Training, PyTorch Distributed).
• Proficiency in data preprocessing, tokenization, embeddings, and language modeling.
• Must be based in the United States.
• Excellent communication and collaboration skills for working with diverse teams.

Culture & Benefits

• Comprehensive medical, dental, and vision insurance.
• 401(k) program with employer match.
• Generous time off for vacation and parental leave.
• Equity in the form of restricted stock units.
• Flexible work environment with remote-first culture and optional office access.