- Zdalnie
- Hybrydowo
... inference runtime (ONNX, TensorRT, TFLite, OpenVINO) and the practicalrealities of running models on-device; hardware-accelerated inference on NPUs/GPUs (Jetson,Qualcomm, Hailo, Ambarella); GenAI/LLM tooling, RAG pipelines, or ML infrastructure; multi-arch andembedded Linux builds (Yocto); a major cloud provider (AWS, GCP, Azure); ...
Poland, Kraków