A proposito di Zhongkai Fu
Hi, I’m Zhongkai Fu, an AI engineer and applied scientist with more than 18 years of experience in natural language processing, machine learning, and AI system development.
I’m an active open-source developer. I created TensorSharp, a native .NET inference engine and agent runtime for running GGUF-based language and multimodal models locally across Windows, macOS, Linux, and mobile devices.
I have also developed projects such as Seq2SeqSharp and CRFSharp, covering neural network training, machine translation, sequence modeling, and traditional NLP.
My work spans the full AI stack—from model training, inference optimization, quantization, GPU acceleration, and distributed execution to tool use, multimodal applications, and agentic workflows.
I have published research at ACL, hold two granted U.S. patents in AI-powered medical summarization, and have written two books about building LLM inference engines, multimodal systems, and agentic runtimes from scratch.
What I enjoy most is connecting research, low-level systems engineering, and real-world product development—turning advanced AI ideas into efficient, reliable, and practical applications.
I’m an active open-source developer. I created TensorSharp, a native .NET inference engine and agent runtime for running GGUF-based language and multimodal models locally across Windows, macOS, Linux, and mobile devices.
I have also developed projects such as Seq2SeqSharp and CRFSharp, covering neural network training, machine translation, sequence modeling, and traditional NLP.
My work spans the full AI stack—from model training, inference optimization, quantization, GPU acceleration, and distributed execution to tool use, multimodal applications, and agentic workflows.
I have published research at ACL, hold two granted U.S. patents in AI-powered medical summarization, and have written two books about building LLM inference engines, multimodal systems, and agentic runtimes from scratch.
What I enjoy most is connecting research, low-level systems engineering, and real-world product development—turning advanced AI ideas into efficient, reliable, and practical applications.
Sostenitori recenti