> Markdown version of [/jobs/ext/4619198-software-principal-engineer-ai-compiler](https://www.wearedevelopers.com/jobs/ext/4619198-software-principal-engineer-ai-compiler). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md).

---

# Software Principal Engineer- AI Compiler

- **Company:** Ampere LLC
- **Location:** Santa Clara, CA, United States
- **Contract:** Permanent contract
- **Skills:** Artificial Intelligence, Artificial Neural Networks, C++ (Programming Language), Data Structures, Python (Programming Language), Pattern Recognition, Pytorch, Deep Learning, Information Technology, SGLang, Hardware Acceleration, llama.cpp, MLIR (Multi-Level Intermediate Representation)
- **Published:** November 7, 2026
- **Apply:** https://www.juju.com/job/28_25303_15c43efd02e5b147946569932955c914

## About the Role

* Bachelors degree in Computer Science, Mathematics or a related technical field & 8 years of related experience; or Master's degree & 6 years
 * Strong CS fundamentals: algorithms, data structures, systems
 * Solid graph algorithm knowledge and reasoning ability
 * Proficiency in Python and C/C++
 * Demonstrated exceptional problem-solving ability - IOI medal, ACM ICPC medal, Codeforces Grandmaster, USACO Platinum, or equivalent competitive programming achievement is a big plus
 * Preferred to have traditional CPU or GPU compiler development background
 * Familiar with LLVM compiler framework as well as MLIR dialects is a plus
 * Fast learner who can pick up new domains quickly and maximize agent-assisted development
 * Familiarity with deep learning concepts and neural network architectures is a plus

## Description

In this role, you will optimize computational graphs to unlock the full potential of Ampere's deep learning accelerator. You'll work across the full SW/HW stack - from inference serving and framework integration down to compiler, runtime, and compute kernels., * 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