Senior / Engineer – AI Model Enablement & Software Tools
Core
Enable next-generation AI models (LLMs, vision, multimodal) on Snapdragon platforms using Qualcomm's AI software stack and model-enablement frameworks.
Role type
Senior IC machine-learning engineer (model enablement & optimization)
Builds
Reusable model support, quantization recipes, and pipeline stages integrated into Qualcomm's SDKs
Domain
Mobile/Edge AI, Systems Software, Hardware-Accelerated Inference
Deliverable
production ML models
Required skills
Python, C/C++, software engineering fundamentals, AI/ML inference concepts, agentic coding tools, debugging, root-cause analysis
Preferred skills
PyTorch, ONNX, Hugging Face, graph-based representations (ONNX Runtime, TorchScript, MLIR), model quantization/calibration, transformer architectures, Linux, embedded/mobile software, Android/RTOS, CMake, Git, agile practices
Technologies
PyTorch, ONNX, Hugging Face Transformers, ONNX Runtime, TorchScript, MLIR, ExecuTorch, TFLite/LiteRT, Snapdragon SoCs, NPUs
Responsibilities
Contribute to bring-up and enablement of new language, vision, and multimodal models on Snapdragon platforms; Study model implementations and computational graphs to understand inputs, outputs, operators, and execution behavior; Work with AI Research and backend teams to identify unsupported operators and performance limitations; Implement model re-authoring, operator decomposition, graph transformations, and backend-specific adaptations; Develop reusable model support for integration into Qualcomm's frameworks; Convert, optimize, and export models from PyTorch and ONNX for execution on Qualcomm AI accelerators; Apply and validate quantization and calibration flows (PTQ, QAT, mixed precision); Measure and improve model accuracy, latency, throughput, memory, and power efficiency; Debug issues across model code, graphs, compilers, runtimes, OS, and hardware layers; Develop unit, integration, accuracy, performance, and on-device validation tests; Write high-quality, maintainable Python and C/C++ code.
Seniority
Senior, hands-on IC
