从React Native + Node.js（4年经验）转型为智能体AI——需要路线图建议
英文摘要
A developer with 4 years of experience in React Native and Node.js seeks advice on transitioning into Agentic AI roles in India. They have completed an AI/LLM course covering Pydantic, LLM theory, API integration, prompt engineering, and local LLMs. They understand the concept of AI agents and RAG, and plan to build projects like a web-search agent and a MongoDB-connected RAG agent. The post asks for feedback on their foundation, a 3-6 month learning roadmap, framework prioritization, portfolio projects, and community resources.
中文摘要
一位拥有4年React Native和Node.js经验的开发者寻求转型为印度智能体AI工程师的建议。他已完成了AI/LLM课程，涵盖Pydantic、LLM理论、API集成、提示工程和本地LLM。他理解智能体AI和RAG的概念，并计划构建如网络搜索智能体和连接MongoDB的RAG智能体等项目。该帖子询问其基础是否扎实、3-6个月的学习路线图、框架优先级、作品集项目以及社区资源。
关键要点
Current background: 4 YOE in React Native/Node.js, comfortable with production APIs, MongoDB, authentication.
当前背景：4年React Native/Node.js经验，熟悉生产级API、MongoDB和身份验证。
Recent learning: Completed an AI/LLM course covering Pydantic, LLM theory, OpenAI/Gemini APIs, prompt engineering, local LLMs via Ollama, and basic agentic AI.
近期学习：完成了AI/LLM课程，涵盖Pydantic、LLM理论、OpenAI/Gemini API、提示工程、通过Ollama的本地LLM以及基础智能体AI。
Planned projects: Web search + calculator agent, MongoDB RAG agent, FastAPI backend with React frontend.
计划项目：网络搜索+计算器智能体、MongoDB RAG智能体、FastAPI后端搭配React前端。
Specific questions about roadmap, framework priority (raw API vs LangChain), portfolio targeting ₹20-35 LPA roles.
关于路线图、框架优先级（原始API vs LangChain）、瞄准₹20-35万卢比年薪的作品集的具体问题。