Jianwen Y.

Hello! I am Jianwen, a skilled Python developer focusing on Large Language Models, RAG, AI Agents, and most importantly, integrating these cutting-edge technologies into your specific use case for real-world outcomes. Whether your goal is for personal use, research, or business revenue, I am glad to help. Skillsets: Chatbot development: LLM-based chatbots using SOTA models like GPT-4.1 and Gemini 2.5 Flash, multi-modal chat, chat with your data: upload your PDFs, text files, audios, images, videos, and query the chatbot about these data. Chatbots that sound like HUMANS. Retrieval Augmented Generation (RAG): Text-based RAG, multi-modal RAG (image, audio, video), advanced RAG with query expansion. And more broadly, information retrieval, search and recommendation. LLM-based agents: LangChain, LangGraph, LlamaIndex, context management, stateful AI with custom memory architectures, chatbots with ability to remember user preferences/customize personality. Structured output & function calling, MCP to query databases and use your browser, and multi-agent orchestration. Web development: HTML, CSS, React.js, FastAPI, JWT Authentication, deployment on AWS or other cloud-based platforms. Large language models: LLM training, fine-tuning, instruction-tuning, reward model training with RLHF/RLAIF, and reinforcement fine-tuning with PPO/DPO/GRPO/REINFORCE++. Speech-to-Text(STT) and Text-to-Speech(TTS) with elevenlabs, openai whisper, and gemini multi-modal models. Training, fine-tuning, and evaluating deep learning models with PyTorch, and performing complex data manipulation and analysis. Other: I am particularly familiar with psychology/psychotherapy and wellness, also academic writing and research.