Core Research Team

The TKLLM team is a collective of interdisciplinary experts and visionary minds, dedicated to pioneering the next generation of large-scale generative AI models. We bring together world-class talent in natural language processing, machine learning engineering, data science, and algorithmic research. Our shared mission is to translate cutting-edge theoretical breakthroughs into powerful, real-world applications that redefine what's possible in domains like text generation and image synthesis.

Our Interdisciplinary Expertise:

• NLP Scientists & Engineers: They are the architects of our models' linguistic intelligence. Their work goes beyond training—it's about teaching our models to understand context, reason with nuance, and generate human-like text. They specialize in advanced techniques like instruction tuning and reinforcement learning from human feedback to ensure our outputs are not just coherent, but also insightful and aligned with user intent.
• ML Infrastructure Engineers: These are the master builders who turn groundbreaking algorithms into robust, scalable systems. They architect the high-performance computing environments that enable the training of massive models, optimizing every aspect of the pipeline for speed and efficiency. Their expertise ensures that our research can be deployed reliably at scale.
• Data Scientists & Engineers: They are the curators of our models' foundational knowledge. Recognizing that data quality is paramount, they design sophisticated pipelines to source, clean, and structure vast multi-modal datasets. Their meticulous work in data preparation and analysis is what allows our models to learn with unparalleled accuracy and depth.
• Algorithm Researchers: As our forward-looking pioneers, they constantly push the boundaries of generative AI. They explore and innovate on the latest architectures—from novel Transformer designs to advanced diffusion models—ensuring that our technology stack remains at the absolute forefront of the field.

Our Collaborative Edge:

What truly defines us is our deeply integrated, collaborative culture. We operate not as siloed functions, but as a unified "research-engineering-data" triad. This creates a powerful innovation cycle where new theoretical insights are rapidly prototyped, engineered into production, and refined through data-driven analysis.

Our approach is fundamentally application-oriented. Every line of code and every algorithm is developed with a clear purpose: to solve complex challenges and deliver tangible value. Whether creating sophisticated conversational agents or enabling controllable image synthesis from abstract concepts, we are committed to building AI that is not only powerful but also practical and accessible.

The TKLLM team is more than just a group of experts; we are builders on a mission to unlock the full potential of generative AI and shape an intelligent future.

Chuanyou Li

Southeast University · PhD/Professor

Chief Scientist

Research Focus: Artificial Intelligence & Machine Learning

Key Contributions: Published numerous papers in top-tier journals, led fundamental algorithm research for AI large models, specializing in AIGC, high-performance computing acceleration, combinatorial optimization, and LLM system architecture.

Kehua Chen

University of Electronic Science and Technology of China · PhD

AI Agent Development Engineer

Research Focus: Government Information Systems

Key Contributions: Contributed to national key research projects, driving AI innovation in government system optimization and implementation.

Guoxiong Tang

University of Pennsylvania · PhD

Deep Learning Engineer

Research Focus: Machine Learning

Key Contributions: Developed high-efficiency machine learning models and algorithms that significantly improved computational efficiency and accuracy in complex data processing.

Wenxin Zhang

University of Rennes · PhD

Professor, Chinese Academy of Sciences

LLM Product Director

Research Focus: Modern Management Theory

Key Contributions: Pioneered streaming engineering design management methodologies that became driving forces for regional economic development.

Ningyu Liang

University of Glasgow · MSc

Data Analysis Engineer

Research Focus: Data Mining

Key Contributions: Developed a high-performance data processing system for efficient handling of large-scale heterogeneous data sources.

Sida Xing

Deakin University · MSc

LLM Algorithm Engineer

Research Focus: Recurrent Neural Networks (RNN)

Key Contributions: Developed Enhanced LSTM (E-LSTM) to address gradient issues in traditional RNNs, improving training efficiency across multiple domains.

Chengjia Wu

Harbin Institute of Technology(ShenZhen) · BC

Hong Kong University of Science and Technology· RA

Director of Advanced Technology

Research Focus: Complex Urban Systems Based on Geospatial Data

Key Contributions:Developing interdisciplinary approaches integrating geospatial analysis, LLMs and complex system theory . Investigating urban spatial structures and socioeconomic interactions through multi-source geospatial data.

Chengyou Xin

University of Electronic Science and Technology of China · MSc

Founder

Research Focus: Algorithm Design and Analysis

Key Contributions:Developed an adaptive learning rate strategy based on Bayesian optimization, reducing convergence cycles for billion-parameter models by 55% (ICML 2024). Built LoRA-XL architecture enabling parallel fine-tuning of thousands of tasks, achieving 95% GPU utilization and processing thousands of fine-tuning jobs daily.