- Cambridge, Massachusetts, United States +1 more
- Full-time
- $148K–$240K per year
- 4+ years
1–2 jobs· newest first
$148K–$240KFull-time
CA$115K–CA$162KHybrid · Full-time · Entry
+10K new in the last hour
Lead performance optimization for large language model training and inference on NVIDIA GPU clusters, improving GPU utilization, memory footprint, and distributed training efficiency.
Expected Base Salary Range $148,000—$240,000 USD
Design, develop, test, and optimize compiler middle-end and backend features for Intel CPU and GPU architectures, focusing on LLVM- and TileIR-based DSL compiler stack.
CAD 114,890.00-162,190.00