🛠 Tech Stack

💼 About This Role

You'll optimize machine learning models for performance on GPU architectures at Roblox, supporting hundreds of ML use cases and billions of inferences daily. You'll conduct low-level profiling, develop best practices, and collaborate across teams to integrate models into production. This role offers the chance to impact a platform connecting tens of millions of users worldwide.

🎯 What You'll Do

• Optimize ML models for performance on GPU architectures.
• Conduct low-level performance profiling and analysis.
• Develop best practices and tooling for model optimization.
• Collaborate with cross-functional teams for integration and deployment.

📋 Requirements

• 6+ years of professional experience and system design expertise.
• Experience debugging GPUs (profiling, Xid errors).
• Proficient in CUDA, Triton, TensorRT for model execution.
• Experience with LLM optimization techniques (speculative decoding, quantization).

✨ Nice to Have

• Passion for building generalized tools for model performance.
• Ability to support internal partners like data scientists.

🎁 Benefits & Perks

• 💰 Equity compensation
• 🏖️ Flexible remote days (Monday/Friday optional in office)
• 🏥 Comprehensive benefits (as per company page)

📨 Hiring Process

Estimated timeline: 2-4 weeks · AI estimate

1. 1Recruiter Call· 30 min
2. 2Technical Screen· 60 min
3. 3Onsite Interviews· 4-5 hours