Avatar

Ziyang Lin

AI Software Development Engineer

Shenzhen Zhiruantong Technology Co., Ltd.

Biography

I am an AI Software Development Engineer at Shenzhen Zhiruantong Technology Co., Ltd., focusing on RAG technology applications and real-time conversational agent development based on multimodal speech models. I graduated with a BSc in Computer Science & Artificial Intelligence from the University of Sheffield in July 2022, where I completed my final year dissertation project on “GLOM for Automatic Speech Recognition” (designing and developing a representation learning model with better interpretability in the field of automatic speech recognition by referencing the concept of GLOM architecture) under the supervision of Dr. Anton Ragni.

Interests

  • Large Language Models & Application Development
  • Retrieval-Augmented Generation (RAG)
  • Multimodal Learning
  • Speech Processing
  • Machine Learning Fundamentals

Education

  • BSc in Computer Science & Artificial Intelligence, 2018-2022

    University of Sheffield

Recent Posts

LLM Agent Multi-Turn Dialogue: Architecture Design and Implementation Strategies

This article provides an in-depth analysis of the core challenges faced by LLM Agents in multi-turn dialogues, detailing the technical …
Last updated on Jul 1, 2025 15 min read Technical Documentation, Artificial Intelligence

Retrieval-Augmented Generation (RAG): A Comprehensive Technical Analysis

This article provides an in-depth analysis of Retrieval-Augmented Generation (RAG) technology, from core architecture to advanced …
Last updated on Jul 1, 2025 28 min read Technical Documentation, Artificial Intelligence

Model Context Protocol (MCP): A Standardized Framework for AI Capability Extension

This article provides an in-depth analysis of the Model Context Protocol (MCP), its core architecture, communication mechanisms, and …
Last updated on Jun 30, 2025 13 min read Technical Documentation, Artificial Intelligence

LLM Tool Calling: The Key Technology Breaking AI Capability Boundaries

This article provides an in-depth analysis of LLM tool calling's core principles, technical implementation, code examples, and best …
Last updated on Jun 30, 2025 15 min read Technical Documentation, Artificial Intelligence

TensorRT In-Depth: High-Performance Deep Learning Inference Engine

This article provides a comprehensive overview of NVIDIA TensorRT's core concepts, key features, workflow, and TensorRT-LLM, helping …
Last updated on Jun 30, 2025 12 min read Technical Documentation, Artificial Intelligence

RAG Data Augmentation Techniques: Key Methods for Bridging the Semantic Gap

This article provides an in-depth analysis of data augmentation and generalization techniques in RAG systems, detailing how to leverage …
Last updated on Jun 29, 2025 11 min read Technical Documentation, Artificial Intelligence

SIP and VoIP Communication Technology: A Comprehensive Guide from Principles to Practice

This article provides an in-depth analysis of SIP protocol and VoIP technology core principles, key components, and implementation …
Last updated on Jun 29, 2025 29 min read Technical Documentation, Network Communication

Modern ASR Technology Analysis: From Traditional Models to LLM-Driven New Paradigms

This article provides an in-depth analysis of modern Automatic Speech Recognition (ASR) technology trends, comparing the design …
Last updated on Jun 29, 2025 6 min read Technical Documentation, Speech Processing

Modern TTS Architecture Comparison: In-Depth Analysis of Ten Speech Synthesis Models

This article provides a comparative analysis of ten modern TTS model architectures, examining their design philosophies, technical …
Last updated on Jun 28, 2025 24 min read Technical Documentation, Speech Processing

Speech Synthesis Evolution: From Traditional TTS to Multimodal Voice Models

This article explores the evolution of speech synthesis technology, from the limitations of traditional TTS models to the integration …
Last updated on Jun 28, 2025 10 min read Technical Documentation, Speech Processing
See all posts

Projects

2020 Assessing the Funniness of Edited News Headlines

This project aims to develop potential solutions for assessing the funniness of edited news headlines

Experience

 
 
 
 
 

AI Software Development Engineer

Shenzhen Zhiruantong Technology Co., Ltd.

Mar 2024 – Present Shenzhen
Responsibilities include:

  • Designing and developing intelligent systems based on Retrieval-Augmented Generation (RAG) technology
  • Participating in advanced Prompt Engineering practices to enhance LLM application effectiveness
  • Developing financial data analysis and government affairs Q&A agents
  • Participating in the company's proprietary multimodal low-latency real-time voice conversation agent
  • Optimizing deployment of large language models and embedding models based on vLLM and llama.cpp
 
 
 
 
 

Software Development Engineer

Shenzhen Institute of Metrology and Quality Inspection

Nov 2022 – Mar 2024 Shenzhen
Responsibilities include:

  • Developing an electric vehicle charging station verification platform website
  • Development of CT online inspection of electronic components project
  • Developing industrial camera applications based on computer vision technologies
  • Participating in cross-platform application development based on the Qt framework
 
 
 
 
 

Event Officer

University of Sheffield Artificial Intelligence Society

Sep 2020 – Jul 2021 Sheffield, United Kingdom
Responsible for hosting and organizing AI-related learning projects (planning workshops or alumni lectures) during the academic term.
 
 
 
 
 

Summer Research Student

Imperial College London

Jun 2020 – Jul 2020 Online
Participated in the 2020 NLP Online Research Summer Project guided by Professor Lucia Specia, developing regression and classification models to evaluate the humor level of edited news headlines.

Domain Knowledge

Large Language Models

Transformer architecture, attention mechanisms, self-supervised learning, fine-tuning techniques

Prompt Engineering

Few-shot Learning, Chain-of-Thought, ReAct framework, prompt template design

Retrieval-Augmented Generation

Vector retrieval, hybrid retrieval strategies, document chunking, context optimization

Multimodal Learning

Cross-modal information fusion, multimodal representation learning, alignment techniques

Machine Learning Fundamentals

Deep learning, reinforcement learning, natural language processing, computer vision

Mathematical Foundations

Linear algebra, multivariate calculus, probability theory

Popular Topics

AI Application Development AI Architecture API API Integration Audio Encoders Deep Learning Large Language Models llama.cpp Local Deployment Model Deployment Model Inference Multimodal Learning Natural Language Processing RAG Real-time Communication Retrieval-Augmented Generation Speech Synthesis Text-to-Speech Tool Calling vLLM

Contact