听说人人都想转型AI工程师? Software to AI Engineer跃迁营来啦!-夜雨聆风

听说人人都想转型AI工程师? Software to AI Engineer跃迁营来啦!

2026，每个IT人都在策划自己在AI时代的职业跃迁。AI Engineer，成为大家最理想的职业方向之一。作为高门槛岗位，你真的了解它所需的真正技能吗？

2026 AI Engineer Bootcamp，带你从软件工程一路实践到AI工程。

AI Eng 001

Software to AI Engineer实践营

核心理念

成为AI时代架构的搭建者！

From Software to AI Engineer, innovate with AI！

项目简介

以企业级全栈教育系统为主线，帮助学生深入理解大模型落地应用与后端微服务架构的工作内容，并亲手完成一个从增强RAG知识检索、Agent智能体开发、云端架构部署（AWS）到自动化运维的完整AI工程项目。

结合AI大模型趋势，带学生利用 LangGraph与MCP解决幻觉并调用后端工具；全流程跑通大厂敏捷研发与运维标准（Jira/PR/CI/CD），告别单纯的API调用时代，真正掌握从RAG 调优、云端部署到自动化运维的工业级全生命周期。

实践是科技唯一目的

AI Eng 002

四大惊喜收获

第一层｜认知革新

学员将掌握AI领域的核心知识与技能：

Software Development Life Cycle（SDLC）

DevOps

AI Agent

Prompt Engineering

Retrieval-Augmented Generation (RAG)

第二层｜工具技能

学员将实际使用企业常见工程与AI工具：

Git
Springboot

Java
AWS

FastAPI

Python
Docker

Terraform

Postgres

LangChain/LangGraph

Jira

Bitbucket

Confluence

第三层｜项目成果

学员将在一套真实完整的工业级全栈架构之上, 用Python独立构建并部署一整层AI服务。项目成果既能拆成一条Fullstack / Backend经历，也能拆成一条AI Engineer经历，一份项目支撑两类岗位投递。

学员最终会带走

✅前后端完整的教育系统架构：前端React (JS / TS), 后端SpingBoot (Java)

✅RAG选课问答助手：基于课程手册的检索增强问答服务，能带依据回答、检索不到时不编造。

✅多工具AI Agent + MCP Server：能查课、选课、答疑,带记忆和多轮对话,通过 MCP 操作真实后端。

✅ 部署上线的 AI 服务：Docker 化、Terraform 部署到 AWS ECS、Bitbucket CI/CD 流水线、接入日志与监控。

第四层｜简历面试

实践营项目完成后，学员可将项目写入简历：

AI-Powered Course Enrolment Assistant | AI Engineer

Tech Stack: Python, FastAPI, LangChain/LangGraph, MCP, Vector DB, Docker, Terraform, AWS (ECS/ECR), Bitbucket CI/CD

Built a production RAG pipeline (chunking → embedding → vector retrieval → context assembly) over course documents, reducing hallucinated answers via source-grounded prompting and a faithfulness evaluation set.

Engineered a multi-tool AI Agent with LangGraph and a custom MCP server, enabling the LLM to query, enrol and advise against a live course-enrolment backend through standardised tool calls.

Containerised the AI service with Docker, provisioned AWS ECS/ECR via Terraform (IaC), and set up a Bitbucket CI/CD pipeline with automated tests and deployment.

Designed a RESTful API layer between the AI service and the existing Spring Boot backend, following layered architecture and TDD practices.

在面试中清晰介绍项目：

示例回答

“I created an AI-powered course selection assistant, which is hosted on a real full-stack course selection system. I wrote a separate AI service layer in Python: first, I created a RAG question-answering tool based on course documents to address the issues of large models being poorly formatted and lacking proprietary knowledge; then, I used LangGraph and MCP to upgrade it into an agent that can actually call the backend to search for and select courses; finally, I deployed it to AWS using Docker + Terraform, and configured CI/CD and monitoring. Throughout the process, we followed the workflow of a real company—Jira managed the tasks, Bitbucket submitted PRs, and I even had on-call rotations. Therefore, I not only know how to call APIs, but I also understand how to adjust the quality of RAG retrieval, how to select tools for the agent, and how to deploy and maintain the service.”

AI Eng 003

授课导师

James – AI | Software | Algorithm Engineer

浙江大学信息工程本科，哥伦比亚大学电子工程硕士。一人AI企业实践者，前Atlassian AI Agent工程师，前阿里算法工师。12年后端开发经验，擅长AI Engineer，后端开发，DevOps。曾斩获Canva，Atlassian，Macquarie，支付宝，携程，IBM等国内外大小公司Offer。

专业背景

12年后端开发经验

澳洲Atlassian AI Agent团队工程师

参与过百万级与千万级用户的大型项目

教学经验丰富，辅导学院斩获Canva, Seek, Quantium等企业Offer