AI Compiler Engineer: Graph Optimizations & HW/SW Co-Design

Ampere Computing; Santa Clara (CA); Hybrid; USD 195,000 - 292,000; Full time

Benefits offered by this job

Health insurance

401K retirement plan

Unlimited flextime

Paid holidays

Job summary

Ampere Computing is seeking a Software Principal Engineer- AI Compiler in California to optimize DL graphs for our energy-efficient AI accelerator. You will work across the SW/HW stack—from inference serving and framework integration to compiler, runtime, and compute kernels—to unlock performance.

You’ll enable PyTorch and Llama.cpp models, implement graph-level optimizations like fusion, and collaborate on HW/SW co-design to push efficiency on Ampere's hardware. 5+ years of exp preferred.

Qualifications

• Strong CS fundamentals: algorithms, data structures, systems.
• Proficiency in Python and C/C++.
• Experience optimizing deep learning graphs and compilers is a plus.
• Competitive programming achievements (IOI/ACM/USACO) are a big plus.

Responsibilities

• Optimize deep learning computational graphs for performance, throughput, and latency on Ampere's accelerator hardware.
• Enable popular models and frameworks (PyTorch, Llama.cpp) and serving platforms (vLLM, SGLang).
• Identify and implement graph-level optimizations: op fusion, pattern recognition, redundancy elimination.
• Collaborate on HW/SW co-design to push computational efficiency.
• Work with cross-functional teams to integrate AI solutions into Ampere's AI hardware platforms.

Skills

Python

C/C++

Algorithms

Data structures

Systems

Competitive programming

MLIR

Education

Bachelor's degree in Computer Science or Mathematics

Master's degree

Tools

LLVM

Job description

Ampere Computing is seeking a Software Principal Engineer- AI Compiler in California to optimize DL graphs for our energy-efficient AI accelerator. You will work across the SW/HW stack—from inference serving and framework integration to compiler, runtime, and compute kernels—to unlock performance.

You’ll enable PyTorch and Llama.cpp models, implement graph-level optimizations like fusion, and collaborate on HW/SW co-design to push efficiency on Ampere's hardware. 5+ years of exp preferred.