蔚来
实习实习-AI应用开发工程师
上海 · 实习 · 2027 届 · 不限
长期有效
任职要求
- Currently pursuing an undergraduate or graduate degree in computer science, software engineering, artificial intelligence, or a related field.
- Familiar with Python, and have basic capabilities in back-end development, API calling and data processing.
- Candidates with knowledge of large language models, AI Agents or RAG, and development experience with frameworks such as LangChain and LangGraph are preferred.
- Familiar with one or more of basic development tools such as Git, Docker, FastAPI and databases.
- Possess excellent skills in information organization, logical expression and Slides production, and be capable of independently completing basic reporting materials.
- Possesses strong learning ability, sense of responsibility and teamwork spirit, and is willing to proactively explore and practice the latest AI technologies.
- Fluent in both Chinese and English, with the ability to proficiently read English technical documents and conduct written communication.
- Candidates who can complete an internship of 6 months or longer will be given priority.
岗位描述
- 1.Participated in the development and implementation of AI Agent projects, covering workflow orchestration, RAG Knowledge Base, tool calling, data processing and application integration.
- 2.Assist in completing the prototype development, functional testing, effect evaluation and problem optimization of the Agent.
- 3.Research and practice relevant technologies such as large language models, AI Agent, and RAG to support the verification of the team's technical solutions and the implementation of functions.
- 4.Assist in organizing project materials, technical proposals and presentation slides, and translate complex technical content into clear and easy-to-understand expressions.
- 5.Complete other project support tasks assigned by the team.
你的简历匹配这个岗位吗？
上传简历，AI 会对照这份 JD 指出你的差距和需要补强的地方
该岗位可能会问的问题
基于 算法 岗位的常见面试问题整理
请解释 Transformer 模型的自注意力机制（Self-Attention）
查看答题思路
1. Self-Attention = 序列中每个 token 与所有 token 计算相关性权重 2. Q=Wq·x, K=Wk·x, V=Wv·x; Attention(Q,K,V)=softmax(QK^T/√d_k)V 3. 除以 √d_k 防止点积过大导致 softmax 梯度消失 4. 多头注意力：多组 QKV 并行，拼接后线性变换，捕获不同子空间特征 5. 对比 RNN：可并行计算，捕获长距离依赖
你做过的最有深度的算法/模型项目是什么
查看答题思路
1. 一句话说清任务：分类/生成/检索/推荐 2. 模型选型：为什么选这个而非那个 3. 数据：数据量、来源、清洗策略 4. 创新点：你做的改进是什么（不是调参） 5. 结果：核心指标提升（准确率/召回/延迟）
过拟合是什么？怎么判断和解决
查看答题思路
1. 过拟合=训练集表现好、验证集表现差，模型记住了噪声 2. 判断：train loss 持续下降而 val loss 上升 3. 解决：更多数据 / 正则化(L1/L2) / Dropout / 早停 / 数据增强 / 降低模型复杂度
解释梯度下降的原理，SGD 和 Adam 的区别
查看答题思路
1. 梯度下降：沿负梯度方向更新参数，使损失降低 2. SGD：每次用随机小批量估算梯度，引入随机性帮助跳出局部最优 3. Adam = Momentum + RMSProp：自适应学习率 + 动量加速 4. Adam 收敛快但可能泛化不如 SGD；CV 常用 SGD，NLP 多用 Adam
给你 10 万条未标注的文本数据，如何找出其中 10 个主题
查看答题思路
1. 预处理：分词、去停用词 2. 向量化：TF-IDF 或 Sentence Embedding 3. 聚类：K-Means（需指定 K=10）或 HDBSCAN（自动确定类数） 4. 或 LDA 主题模型直接输出主题-词分布 5. 评估：看每个簇的高频词判断主题合理性，迭代调整
更多面试题库功能即将上线
信息来源：企业官方招聘页。投递前请以官网信息为准。