Hire the Best Artificial Intelligence Engineers
in China

I build production-ready AI Agent systems designed to deliver predictable, debuggable, and auditable outcomes, not demos or prompt experiments. My work focuses on deterministic workflows, multi-agent orchestration, memory management, and hallucination control, ensuring AI behavior is reliable in real-world use. I help companies and founders turn AI into operational systems and market-ready products, with end-to-end delivery from architecture and workflow design to RAG, validation, and monitoring. My approach prioritizes failure detection, recoverability, and observability, so systems are maintainable and scalable over time. Where relevant, I integrate full-stack web solutions to connect AI capabilities with applications, and I have experience enabling products for the Chinese market, including localization-aware design and market-aligned execution. I work best with teams or founders who need highly reliable AI systems, production-ready workflows, and solutions that reduce risk while enabling business outcomes.

Education: Master of Science in Computer Science, Stanford University Working Background: 12+ years full-stack & Generative AI engineering experience, 7 years specialized in LLM application, RAG pipeline, agentic workflow, enterprise AI governance & conversational bot development; long-term freelance & contract AI technical consultant serving North American MSP, SaaS, training & professional service clients. Core Professional Summary A Stanford CS graduate with deep foundational computer science theory and end-to-end production AI delivery capabilities. I specialize in building scalable, low-hallucination LLM systems, autonomous AI agents via LangChain & LangGraph, enterprise RAG knowledge platforms, multi-channel conversational bots (voice bot, WhatsApp business chatbot), and Microsoft Copilot / Anthropic Claude enterprise deployment & governance solutions. I align fully with the technical requirements of all listed job postings: proficient in Python, vector database architecture (Pinecone, Chroma, Weaviate), full RAG pipeline design & hallucination optimization, custom tool agent workflows, third-party API integration (Google Drive API, business line systems, calendar platforms), and production-grade AI deployment with logging, error recovery & data security compliance. Equipped with client-facing consulting skills, capable of translating complex AI technical logic into plain language for non-technical business stakeholders, completing project scoping, timeline planning, risk assessment and staff training independently. Familiar with compliance standards including SOC 2, CMMC, ISO/IEC 42001 AI management systems, HIPAA and TCPA for regulated industry AI deployments. Key Technical Capabilities (Matching All Job Ad Requirements) LLM Orchestration & Agent Development Master of LangChain and LangGraph for building stateful, multi-step agentic workflows; skilled in ReAct logic, tool calling, human-in-the-loop routing, multi-agent collaboration, and task decomposition for complex business objectives. Experienced building custom autonomous agents with self-planning, document retrieval and cross-system data query abilities, capable of constructing lightweight demo agents and enterprise-grade state machine systems. End-to-End Production RAG Engineering & Hallucination Mitigation Full-cycle RAG pipeline construction: document parsing (PDF, Google Docs, Sheets), semantic chunking, embedding generation, hybrid vector search, reranking, metadata filtering, source citation control. Successfully optimized existing RAG systems to cut hallucination rate by over 70% through grounded prompt constraints, retrieval quality tuning and evidence consistency verification. Developed Google Drive synchronized knowledge chatbots that strictly restrict model output to uploaded internal documents without relying on general pre-trained knowledge, supporting automatic file update & source reference display. Vector Database Design & Management Hands-on construction and maintenance of Pinecone, Chroma, Weaviate vector storage clusters; design vector indexing strategies, latency optimization, data isolation and scalable embedding pipelines for knowledge base platforms supporting mass document ingestion. Enterprise LLM Deployment: Microsoft Copilot & Anthropic Claude Deep practical experience administering Microsoft 365 Copilot tenant configurations, policy governance, agent customization and Frontier program tools (Cowork, Scout); full-stack deployment of Claude Team / Claude Code for corporate clients, build custom MCP connectors to link document management, PSA, HR and calendar business systems, design permission scoping & Entra ID identity hardening to control AI data access scope. Conversational AI: Voice Bots & WhatsApp Business Bots Develop and upgrade production-ready voice AI agents (Vapi.ai, ElevenLabs, OpenAI Whisper STT/TTS) with natural dialogue flow, interruption handling, session context retention and calendar integration (Google/Outlook/Calendly); refactor WhatsApp Cloud API bots to strengthen backend stability, add comprehensive logging, fault recovery and multi-turn conversation memory, proficient in Node.js/TypeScript & Python dual-stack backend development for chatbot services. Prompt Engineering, API Integration & AI Governance Expert in few-shot, chain-of-thought, structured JSON output prompt design to standardize LLM response format; wide experience integrating REST/OAuth third-party business APIs for AI workflow automation. Build complete AI governance frameworks covering data exposure mapping, access control, audit logging, complying with enterprise security & AI compliance standards. Development & Delivery Stack Primary language: Advanced Python; Secondary: TypeScript/Node.js Deployment tools: FastAPI, Docker, CI/CD, cloud hosting Frontend interface: Simple web chat widgets, Slack/Discord integration for internal AI tools Representative Project Experience (Fully Cover All Posted Jo Best Wishes Your sinere Wang D.

I am a Senior Unity Client Engineer with over 5 years of experience creating engaging games for mobile and console platforms. My expertise lies in Unity3D, C#, and live-ops event systems, enabling me to enhance player experiences for titles with millions of daily active users. I excel in performance optimization, UI frameworks, and cross-platform adaptation, ensuring smooth gameplay across devices. My approach focuses on clean, maintainable code and effective collaboration across teams, allowing me to contribute to major projects efficiently. I harness AI-assisted development tools to streamline workflows and elevate product quality. If your project requires a skilled developer who understands the intricacies of game development and is committed to delivering exceptional experiences, I would love to discuss how I can help bring your vision to life.

I build AI agents that survive production — LangGraph multi-agent systems, RAG that retrieves the right chunk, and MCP servers on live databases. Recent: a 6-agent booking platform on FastAPI + LangGraph handling flights, hotels and live payments. WHAT I BUILD - Multi-agent systems — LangGraph and LangChain orchestration, state machines, checkpointing, human-in-the-loop gates, multi-LLM routing across OpenAI, Claude and open models - RAG that actually retrieves — chunking strategy, embeddings, hybrid search, reranking, pgvector and Pinecone, plus an eval harness so you can prove retrieval improved instead of guessing - MCP servers — custom Model Context Protocol integrations connecting Claude and other clients to your databases, APIs and internal tools - Voice AI — streaming ASR and TTS over WebSocket, realtime conversational pipelines, built at the protocol level rather than wrapped around a hosted API - Agent + automation glue — n8n, Make and Zapier wired to real tool-calling agents, not just LLM prompt nodes - LLM optimization — prompt engineering, fine-tuning with LoRA/PEFT, and honest advice on when retrieval beats fine-tuning PROOF, NOT CLAIMS - smart_booking_agent — FastAPI + LangGraph platform coordinating six domain agents (flight, hotel, ride, ticket, planning, Q&A) with multi-LLM routing, Redis checkpointing, and live third-party booking and payment integrations including WeChat Pay v3 - xiaozhi-esp32-server — I lead this open-source voice AI assistant running on ESP32 hardware: WebSocket streaming, ASR, LLM dialog, RAG and function calling on a 240MHz chip. Publicly auditable. - Claude MCP server — production integration wiring Claude Code to a live database with scoped tool schemas and auth boundaries. 5-star reviewed on Upwork. - Fintech ML — credit scoring, recommendation and time-series forecasting models on noisy financial data. BACKGROUND M.S. in Pattern Recognition and Intelligent Systems, Beihang University (BUAA), China Scholarship Council scholar. Research: tree-structured Transformers for semantic parsing, outperforming baseline Transformers on three benchmarks. 5+ years shipping Python systems end to end — model to API to deployment. TECH STACK Python · LangGraph · LangChain · LlamaIndex · FastAPI · PyTorch · Hugging Face · OpenAI and Anthropic APIs · MCP · pgvector · Pinecone · Redis · PostgreSQL · MongoDB · Docker · n8n · Make HOW I WORK Based in Shenzhen (UTC+8) — I overlap comfortably with Asia, Australia, the Gulf and European mornings. I scope before I code, I tell you when something won't work, and I hand over documented, tested code you own outright. Send me the problem, not the spec. I'll tell you what it actually takes.

Greetings! I am Chang Liu, an AI/ML engineer and Ph.D. graduate from the University of Science and Technology of China (USTC), specializing in computer vision, diffusion models, and large language models. I focus on bridging the gap between research and robust engineering, helping startups, research labs, and clients to build reliable deep learning projects and publish top-tier papers. The "Deep Expertise" Guarantee: In an era where AI agents like Codex and Claude Code make coding easy, the true bottleneck in deep learning deployment is domain-specific alignment. When you hire me, you get direct access to a published Ph.D. expert who knows exactly how to preprocess your in-domain data, which encoder works best for your project, and what pipeline best satisfies both your performance and novelty needs. No unverified AI-generated code. No junior developers delivering copy-pasted projects. Just precise any-to-any modeling for your exact task. ◆ Why Work With Me? ▪ Ph.D.-trained researcher from USTC with 10+ authored publications (AAAI, ICCV, TIP, TMM). ▪ Earned 2.1k+ total GitHub stars across open-source generative AI and vision repositories (@AlonzoLeeeooo). ▪ Authored 600+ professional blogs on frontier AI, gaining 30,000+ likes on Zhihu and 13,000+ followers across major Chinese platforms (@叫我Alonzo就好了). ▪ Hands-on experience helping multiple startups deploy specialized AI-powered solutions since 2024. ▪ Privately guided 10+ clients to successful publications in NeurIPS, CVPR, ACL, and EMNLP. ◆ How Can I Support Your Project? ＞ Deep Learning Implementation & Problem Solving ▪ Architecture & Alignment: Translating your conceptual ideas into technical documentation and robust code deployments. ▪ End-to-End Execution: Planning experiment iterations tailored to your business needs across key domains: - Generative AI & Diffusion Models: Text-to-X generation, image/video editing, and 3D-related tasks. - Computer Vision Pipelines: High-level vision (segmentation, detection) and low-level vision (processing, restoration). - NLP & LLM-Related Tasks: Text classification, RAG systems, LLM reasoning, and domain-specialized applications like radiology report generation. ＞ Academic Mentoring ▪ Literature Review & Positioning: Conducting comprehensive literature surveys based on your intended research direction to identify gaps and position your work effectively. ▪ Innovation & Experimentation: Proposing technical novelties, preparing rigorous experimental codebases, and tuning experiments to validate your hypotheses. ▪ Writing & Polishing: Structuring compelling scientific narratives, clearly highlighting technical novelty, and executing professional-grade paper polishing & formatting. ▪ Manuscript Reviewing: Providing professional, highly constructive review comments on your submissions. ▪ Top-Tier Rebuttals: Formulating clear strategic responses to challenging reviewer feedback, and thoroughly polishing your rebuttal cover letters for elite venues (e.g., AAAI, NeurIPS, EMNLP, ACL, TIP). ◆ Have a Project or Research Idea in Mind? Every project or mentorship I take on gets my full technical and academic attention. If you have a project in mind or a research paper that needs rescuing, send me your topic, current stage, and target outcome. I am always happy to help. ◆ Where to Learn More about Me? ＞ Profiles & Technical Content: ▪ GitHub: @AlonzoLeeeooo (2.1k+ stars). ▪ Technical Blogs: @叫我Alonzo就好了 across Chinese media platforms (Zhihu, WeChat Official Accounts, Xiaohongshu). ▪ Google Scholar: Chang Liu, USTC (450+ citations). ＞ Representative Publications (11 Selected): - Aspect-Based Sentiment Analysis with Context Denoising (NAACL 2024) - Bootstrapping Large Language Models for Radiology Report Generation (AAAI 2024) - Toward Interactive Image Inpainting via Robust Sketch Refinement (TMM 2024) - Exploiting Diffusion Prior for Real-World Image Dehazing with Unpaired Training (AAAI 2025) - When Schrödinger Bridge Meets Real-World Image Dehazing with Unpaired Training (ICCV 2025) - S2D-Align: Shallow-to-Deep Auxiliary Learning for Anatomically-Grounded Radiology Report Generation (AAAI 2026) - BookAgent: Orchestrating Safety-Aware Visual Narratives via Multi-Agent Cognitive Calibration (ACL 2026) - Physics Meets Perception: A Reinforcement Learning Framework for Unpaired Real-World Image Dehazing (ECCV 2026) - LaCon: Late-Constraint Controllable Visual Generation (TIP 2026) - StableV2V: Stabilizing Shape Consistency in Video-to-Video Editing (TCSVT 2026) - A Systematic Review of Deep Learning-based Research on Radiology Report Generation (Submitted to TPAMI)

AI-Powered Medical Systems Developer | Clinical Decision Support | Dental & Diagnostic AI | 8+ Years Python I am a senior Python developer with over 8 years of professional experience, specializing in AI-driven medical systems, clinical decision support platforms, and healthcare automation solutions. Unlike general developers, I focus specifically on building production-ready medical systems that improve diagnostic accuracy, clinical workflow efficiency, and structured reporting. 🚀 What I Have Built 🔹 AI-Based COVID Detection System Designed and developed an intelligent medical screening system using AI techniques to rapidly analyze clinical indicators and assist in early-stage diagnosis. The system emphasized efficiency, data processing accuracy, and real-world deployment. 🔹 Dental 2D Cephalometric Landmark & Analysis System Independently developed a Python-based dental diagnostic system capable of automated landmark detection and cephalometric analysis. This system supports orthodontic diagnosis and treatment planning by improving measurement accuracy and reducing manual workload. 🔹 Clinical Decision Support Systems (CDSS) Experienced in building systems that: Perform structured risk assessments (e.g., CAMBRA-like models) Generate evidence-based treatment recommendations Automate structured PDF clinical reports Support regulatory update mechanisms Maintain data consistency and compliance awareness 🧠 My Technical Expertise ✔ Python (8+ years) ✔ FastAPI / Django / Flask ✔ AI & Machine Learning (PyTorch / TensorFlow) ✔ Medical data processing & workflow modeling ✔ Automated PDF report generation (ReportLab, WeasyPrint, wkhtmltopdf) ✔ Rule-based engines for clinical logic ✔ PostgreSQL / MySQL / MongoDB ✔ Secure system architecture & compliance-aware design 💡 What I Can Help You With Build or enhance AI-powered clinical decision support systems Develop automated medical report generation modules Implement regulatory or guideline update mechanisms Create diagnostic automation tools Design structured risk assessment engines Develop dental or medical imaging analysis systems Improve performance and architecture of existing healthcare platforms 🎯 Why Clients Choose Me ✔ Deep understanding of clinical workflows (not just code) ✔ Independent system-level development capability ✔ Clean, scalable, maintainable architecture ✔ Strong problem-solving skills ✔ Clear communication & reliable remote collaboration I am particularly interested in long-term collaborations with medical startups, dental software companies, and healthcare SaaS platforms seeking to build reliable, intelligent, and scalable systems. If you're building or upgrading a healthcare AI platform, I would be glad to discuss how I can contribute to your success.