Neural Magic is hiring a
Software Engineer

Summary

The job is for a Software Engineer focused on vLLM at Neural Magic, a series A startup. The role involves writing high-performance code for GPUs, collaborating with machine learning teams, and working on performance analysis and modeling. The company offers competitive compensation, health care, retirement plan, paid time off, family leave, disability coverage, professional development opportunities, flexible work arrangements, wellness resources, and free food.

Requirements

• BS, MS, or PhD in computer science or computer engineering
• Extensive experience in writing high performance code for GPUs
• Deep knowledge of GPU hardware
• Knowledge of CUDA
• Experience with high performance computing
• Knowledge of mathematical software, especially linear algebra or signal processing
• Knowledge of modern C++
• Knowledge of tensor computations
• Knowledge of deep neural network models and techniques

Responsibilities

• Write robust and modern C++ and CUDA, working on high-performance machine learning primitives, performance analysis and modeling, and numerical methods
• Review code and help develop best practices for the team
• Work closely with machine learning teams to attain high performance for neural networks in vLLM

Preferred Qualifications

• A self-directed practitioner who learns fast
• Strong communications skills with both technical and non-technical team members
• Strong sense of project ownership and personal responsibility

Benefits

• Competitive compensation and stock option plan
• Comprehensive health care (medical, dental, vision)
• Retirement plan (401k, IRA)
• Generous paid time off (vacation, sick leave, holidays)
• Family leave (maternity, paternity)
• Disability coverage
• Professional development opportunities
• Flexible work arrangements (remote options)
• Wellness resources
• Free food and snacks (in the office)