Tasks:
- * Adapt and deploy embodied AI models on robot and custom chip platforms
- * Assess deployment feasibility with algorithm and chip teams
- * Build model deployment toolchains and release workflows
- * Convert and quantize models
- * Evaluate latency, memory, power, and compute utilization
- * Optimize edge inference performance
- * Validate model accuracy and automate regression testing
Perks/Benefits:
Skills/Tech stack required:
[Accuracy Validation] [Action models] [C++] [CUDA] [Data Structures] [Data Structures and Algorithms] [Deep learning] [Edge inference] [Edge inference deployment] [Embedded Linux] [Inference deployment] [Language Models] [Llama.cpp] [Mixed Precision] [Mixed precision inference] [MLC LLM] [Model Conversion] [Model export] [Model Quantization] [Multimodal Models] [NPU] [ONNX] [Performance Benchmarking] [Precision Inference] [PTQ] [Python] [PyTorch] [QAT] [TensorRT] [Transformer] [Vision-language] [Vision-language-action] [Vision-Language-Action Models] [Vision Language Models]
Educational requirements:
[Bachelor's Degree]
Role(s):
[AI Deployment Engineer] [AI Infrastructure Engineer] [Deployment Engineer] [Engineer] [Infrastructure Engineer] [Learning Engineer] [Machine Learning Engineer]