Professional Summary
Innovative AI Agent Engineer with expertise in developing intelligent autonomous systems and multi-agent architectures. Specialized in large language models, reinforcement learning, and agent-based frameworks. Proven track record in designing and deploying production-ready AI agents for complex decision-making tasks. Passionate about advancing the frontier of artificial intelligence through practical agent systems that solve real-world problems.
Professional Experience
- Conducted research under the supervision of Prof. Yan LIN and Prof. Hongchuan SHEN
- Developed Financial Agent for automated financial text analysis and model continuous learning
- Built Computer Agent capable of automating computer operations through natural language instructions and screen capture analysis
- Created ShortStudio, an AI-powered short video creation platform that analyzes trending videos and generates similar content
- Developed multimodal emotion analysis system for short drama content
- Fine-tuned Qwen3-8B model for stance detection using Zhihu data, improving accuracy from 0.77 to 0.93
- Open-sourced MSJ-Factory project: A Complete Guide to Fine-tuning Qwen2.5-Coder for Chinese Sentiment Analysis (150+ GitHub stars)
- Conducted research under the supervision of Prof. Dimitrios Kanoulas and Dr. Jianhao Jiao
- Implemented real-time object tracking and position-based planning on Unitree Go2 quadruped robot platform
- Developed end-to-end environmental mapping system integrating predefined knowledge graphs with real-time observations
- Integrated scene graphs, vision-language models, and graph-based retrieval-augmented generation for indoor navigation
- Developed embodied AI system for humanized operations on short video platforms
- Built low-cost unmanned surface vessel equipped with action camera for water photography and 3D reconstruction
- Developed obstacle avoidance system with object recognition and depth estimation capabilities
- Implemented KAN layers to replace MLP layers for feature embedding in iPlanner
- Conducted research under the supervision of Prof. Zeng Jicheng and Prof. Lin Yan
- Collected 330,000 game reviews and utilized large language models for automated game analysis
- Built synthetic game review dataset for training game issue comment checker and released as open-source
- Applied difference-in-differences methodology to study the impact of large language models (e.g., GPT) on HuggingFace
- Conducted music analysis of Suno songs combining music theory, signal processing, and machine learning techniques
- Developed end-to-end TCMFormer Traditional Chinese Medicine diagnostic model based on TCM theory
- Created business report visualization website utilizing commercial data
- Validated multiple machine learning algorithms using public datasets and published research papers
- Collaborated with Prof. Xiao Li's robotics team at Huazhong University of Science and Technology
- Developed autonomous navigation system for 0.8-meter rescue unmanned surface vessel
- Implemented multi-agent reinforcement learning on 1.75-meter unmanned vessel array
Technical Skills
Agent Frameworks:
LangChain, AutoGPT, BabyAGI, CrewAI, LangGraph, Semantic Kernel
LLM & APIs:
GPT, Claude, Gemini, Qwen, Mistral, OpenAI API, Anthropic API
Programming:
Python, TypeScript, JavaScript, C++
ML/DL Frameworks:
PyTorch, TensorFlow, Transformers, TRL
Vector Databases:
Pinecone, Weaviate, Chroma, Qdrant, FAISS
Tools & Platforms:
Docker, Kubernetes, AWS, GCP, FastAPI, Redis, PostgreSQL
Publications
- Qianyi Zhang, Shijian Ma, Boyi Liu, Jianhao Jiao, Dimitrios Kanoulas. "Follow Everything: A Leader-Following and Obstacle Avoidance Framework with Goal-Aware Adaptation." arXiv, 2025.
- Shijian Ma, Shicong Ma, Jianhao Jiao. "RoboAgent: An Integrated Hardware-Software Agent for Short Video Interactions." TechRxiv, 2024.
- Shijian Ma, Shicong Ma, Jianhao Jiao. "WaveDepth: Enabling Water Depth Estimation with Water Segmentation." TechRxiv, 2024.
- Shijian Ma, Shicong Ma, Jianhao Jiao. "WaveShot: A Compact Portable Unmanned Surface Vessel for Dynamic Water Surface Videography and Media Production." arXiv, 2024. AIME accept.
- Shijian Ma, Shicong Ma. "Exploring the Relationship between Vehicle Specifications and Fuel Efficiency using Random Forest Regression and Gradient Boosting Regression Models." ECCS, 2024.
- Shijian Ma. "Churn Prediction in Business using Logistic Regression and Logit Boost." ICDSCA, 2023.
- Shicong Ma, Xiaoguang Tang, Shijian Ma. "Integration of Multi-sensor Marine Environment Monitoring System for Hybrid-Power USV." ICGMRS, 2023.
- Shijian Ma, Xiaoguang Tang, Weize Ma, Jitao Liu, Li Xiao. "An Unmanned Cleaning Robot for the Inner Wall of Large LNG Tanks." ICARM, 2022.
Key Projects
- Developed intelligent agent for automated financial text analysis and processing
- Implemented continuous learning mechanism for model adaptation to evolving financial data
- Built autonomous agent capable of executing computer operations via natural language instructions
- Integrated screen capture analysis for visual understanding and context-aware actions
- Implemented multi-step task planning and execution capabilities
- Developed platform for analyzing trending short videos and generating similar content
- Integrated video analysis algorithms to identify successful content patterns
- Implemented automated content generation pipeline using AI models
- Built comprehensive emotion analysis system for short drama content
- Integrated multimodal analysis combining visual, audio, and textual features
- Developed emotion recognition and classification models for dramatic content
- Created comprehensive guide for fine-tuning Qwen2.5-Coder for Chinese sentiment analysis
- Fine-tuned Qwen3-8B model for stance detection using Zhihu data
- Achieved significant accuracy improvement from 0.77 to 0.93 in stance detection tasks
- Open-sourced project gaining 150+ GitHub stars for community contribution
- Implemented real-time object tracking and position-based planning on Go2 quadruped robot
- Achieved open-set object following capabilities for diverse target objects
- Created maps by integrating predefined knowledge graphs with real-time observations
- Integrated scene graphs, vision-language models, and graph-based RAG
- Applied to indoor navigation and scene understanding tasks
- Developed an autonomous social media interaction system
- Utilized computer vision for precise interface recognition and physical interaction
- Generated contextually relevant content through large language models
- Designed a USV with real-time camera and audio transmission
- Integrated 4G module for remote video streaming
- Combined LLMs with music generation and 3D generation models
- Combined water semantic segmentation with water scene depth estimation
- Constructed specialized water scene depth estimation dataset
- Designed lightweight real-time estimation model optimized for edge devices
- Developed USV-based underwater scene depth estimation system
- Integrated single-beam sonar and RTK-GPS for underwater 3D reconstruction
- Built Nginx-based point cloud visualization server
- Designed lightweight and portable water surface filming unmanned vessel
- Implemented innovative highly maneuverable water surface photography method
- Integrated monocular depth estimation algorithm to enhance spatial awareness
- Designed neural network architecture integrating mirror symmetry and Hodge theory
- Developed network components including Hodge duality layers and symmetry modules
- Achieved superior performance on data with intrinsic symmetry
- Developed tongue diagnosis model based on ResNet and mobile application
- Created TCMFormer model with expanded Traditional Chinese Medicine terminology
- Implemented ALiBi positional encoding and LoRA fine-tuning
- Developed feature selection method based on matroid theory
- Created correlation-based dependency analysis for identifying maximally independent feature sets
- Real Estate Price Analysis: Comprehensive machine learning analysis of property prices
- MPG Prediction: Applied Random Forest and Gradient Boosting regression models
- Student Writing Score Prediction: Implemented SVM and kNN algorithms
- Laptop Price Prediction: Comparative analysis of regression models
- Business Churn Prediction: Applied Logistic Regression and Logit Boost algorithms
- Collaborated on field testing and co-authored conference paper for hybrid-power USV
- Enhanced robot stability through extended buoyancy system redesign
- Implemented modular battery system for easier maintenance and replacement
Honors & Achievements
Professional Service
- Reviewer: IROS 2025, ICRA 2025, RA-L 2024, IROS 2024
Patents
- Patent No: ZL201420713314.1 - A portable remote control water cleaning boat (2022, China)
- Patent No: ZL202130395091.4 - Water surface intelligent cleaning robot (2022, China)
Awards
- Silver Medal (Hubei Province Level), 7th National College Students' "Internet plus" Entrepreneurship Contest (2022, Wuhan, China)
- Winner's Prize Award, 16th "Chunhui Cup" Chinese Overseas Students Innovation and Entrepreneurship Competition (2022, Wuhan, China)
Certificates
- Certificate in Data Analysis Professional Skills, Chinese Association for Applied Statistics (CAAS) (2024, China)
- Research Camp Program: Academic Training, ViaX Instructor (2022, China)