3 days ago
San Diego, CA, USAMid Level / Senior
H1B sponsor
Base Salary
$141k - $211k/yr
Responsibilities
- Enable new language, vision, vision-language, and multimodal models on Snapdragon platforms.
- Analyze computational graphs, model implementations, operators, tensor shapes, data flow, and execution behavior.
- Implement model re-authoring, operator decomposition, graph transformations, layout changes, and backend-specific adaptations.
- Develop reusable model support, pipeline stages, model mappings, quantization recipes, configuration models, artifact handling, and validation utilities.
- Convert, optimize, and export PyTorch and ONNX models for Qualcomm AI accelerator execution.
- Apply and validate quantization and calibration, including weight and activation quantization, mixed precision, PTQ, and accuracy recovery.
- Measure and improve model accuracy, latency, throughput, memory consumption, model size, and power efficiency.
- Work with transformer and generative AI inference concepts including attention, KV cache, prefill, decode, LoRA, MoE, and speculative decoding.
- Debug across model code, graphs, quantization, compilers, runtimes, operating systems, and hardware.
- Develop unit, integration, accuracy, performance, and on-device validation tests.
- Write maintainable Python and C/C++ code and participate in design reviews, code reviews, debugging, and testing.
- Collaborate with research, hardware, QA, product, and customer-facing teams; Senior Engineers also own larger features and mentor junior engineers.
Requirements
- A bachelor’s degree in Computer Science, Computer Engineering, Electrical Engineering, or a related field is required; a master’s degree is preferred.
- The Engineer level requires project or internship experience or 1+ year of software development experience; the Senior Engineer level requires 2+ years of software design, development, and delivery experience.
- Strong programming skills in Python and/or C/C++ and understanding of data structures, algorithms, object-oriented design, debugging, testing, and source control.
- Foundational understanding of neural networks, deep learning, AI/ML inference, and model execution workflows.
- Practical experience using agentic coding tools such as Claude Code, OpenAI Codex, or equivalent tools, including reviewing and validating generated code.
- Ability to understand unfamiliar software and model implementations, perform structured root-cause analysis, and communicate technical concepts clearly.
- Preferred experience includes PyTorch, ONNX, Hugging Face Transformers, graph tooling, model quantization and calibration, transformer models, Linux, embedded or edge software, Android, QNX, CMake, Git, testing, profiling, and AI accelerator architectures.
- Preferred exposure includes Qualcomm AI software stacks, model-enablement frameworks, Snapdragon SoCs, NPUs, HTPs, and model conversion, compilation, deployment, and runtime tools.
Benefits
- Competitive annual discretionary bonus program and potential annual RSU grants.
- Competitive benefits package supporting employees at work, at home, and at play.
- Qualcomm provides reasonable disability accommodations during the application and hiring process.
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.