Intern - GenAI Benchmarking (MLE) / On-Device Model Deployment (SWE)
Core
Intern role focusing on restructuring open-source generative AI models for efficient on-device inference or developing reliable benchmarking frameworks for model evaluation.
Role type
Intern, Machine Learning Engineer / Software Engineer
Builds
Python libraries for model architecture restructuring, quantization, and inference utilities; benchmarking frameworks for generative models
Domain
Generative AI, Edge AI, Model Efficiency, Evaluation Methodology
Deliverable
production ML models
Required skills
PyTorch, Python, Large Language Models, Generative AI systems, Software development, Debugging, Reproducible experiments
Preferred skills
Qualcomm AI Stack, Inference runtimes, Hardware accelerators, Model quantization, Statistical thinking, CI/CD workflows, AI agent frameworks
Technologies
PyTorch, Hugging Face Transformers, Diffusers, AIMET, ONNX Runtime, ExecuTorch, llama.cpp, LangChain, LlamaIndex, Autogen
Responsibilities
Restructure model architectures for hardware constraints; Implement efficient inference techniques; Research and improve model evaluation reliability; Validate quantized models against reference models; Read research papers and implement proposed techniques; Write tested, documented code
Seniority
Intern
