6 months ago
Base Salary
$158k - $238k/yr
Responsibilities
- Architect and deliver model optimization strategies that transform PyTorch models for efficient inference on Qualcomm accelerators.
- Drive graph capture and deployment using PyTorch, ONNX, torch.compile, and related graph-level transformations.
- Design and implement fusion kernels using Triton or similar DSL-based approaches.
- Collaborate with compiler, performance, accuracy, and runtime teams on lowering, kernel fusion, layout, and integration strategies.
- Profile and optimize LLM, VLM, and diffusion inference for throughput and latency across batch sizes, sequence lengths, and serving modes.
- Optimize transformer-specific behavior, including KV cache management, decoding, and long-context performance.
- Enable continuous batching and assess its effects on memory, scheduling, and tail latency.
- Architect distributed inference strategies such as sharding and parallelism across multi-core and multi-device systems.
- Create reusable tooling and optimization patterns for new hardware architectures.
- Debug complex performance and stability issues, identify root causes, and deliver production-ready solutions.
Requirements
- Expert-level PyTorch and inference-focused model optimization experience with strong Python engineering skills.
- Hands-on experience with torch.compile, TorchDynamo, or related graph-capture and compilation workflows.
- Deep understanding of transformer architectures, attention mechanisms, mixture-of-experts models, and performance trade-offs.
- Practical experience with KV cache behavior, serving-time optimization, and memory/performance trade-offs.
- Strong foundation in computer architecture, machine-learning accelerators, and distributed systems.
- Proven ability to lead cross-functional technical efforts and influence design decisions.
- Bachelor's degree in Computer Science, Engineering, Information Systems, or a related field plus 4+ years of related experience; alternatively, a master's degree plus 3+ years or a PhD plus 2+ years.
- Preferred qualifications include Triton or similar DSL fusion-kernel development, collaboration with ML compiler teams, familiarity with LLM serving and continuous-batching systems, and experience with numerical methods, performance/accuracy analysis, or evaluation frameworks.
- A PhD in a relevant field is preferred.
Benefits
- Annual discretionary bonus program and opportunity for annual RSU grants are available.
- Qualcomm offers a competitive benefits package designed to support employees at work, at home, and at play.
- Qualcomm provides reasonable disability accommodations during the application and hiring process.
- The posting lists a pay range of $158,400.00–$237,600.00 for the applicable job location.
Categories
About Qualcomm
Qualcomm is a public semiconductor company headquartered in San Diego, founded in 1985, that designs and sells wireless chipsets and platforms for mobile devices, automotive, IoT, and networking, notably the Snapdragon application processors and 5G modems. It also licenses a large portfolio of cellular patents to device makers, generating revenue alongside chip sales; its technology underpins many Android smartphones and emerging automotive and edge-compute systems, and it trades on NASDAQ as QCOM.