AI工程师
AI Engineer
职位描述
我们正在寻找一名AI工程师，负责设计、构建和扩展驱动我们AI产品的系统。你将在大型语言模型、多智能体系统和生产软件的交汇处工作，将前沿的AI能力转化为可靠、面向用户的产品体验。
职责内容
· 设计并实现基于LLM的流程，包括提示工程、上下文管理和响应合成
· 构建并优化多智能体编排系统，使AI组件能够交互、推理并生成连贯的输出
· 开发并维护与基础模型API（Anthropic、OpenAI等）的集成，管理大规模下的延迟、成本和可靠性
· 实现检索增强生成（RAG）和记忆系统，以支持持久且上下文感知的行为
· 与评估/QA工程师合作，构建衡量输出质量、连贯性和事实依据的框架
· 与产品和设计团队协作，将工作流程转化为稳健的技术体验
· 在整个技术栈中交付生产功能，从模型层到应用层
· 监控、调试并提升系统性能、令牌效率和安全机制
任职要求
· 3年以上软件工程经验，有构建LLM驱动应用的实际经验
· 精通Python和/或TypeScript/JavaScript
· 有LLM API、提示工程和智能体框架（如LangChain、LlamaIndex或自定义编排）的经验
· 熟悉向量数据库和RAG架构（如Pinecone、Weaviate、pgvector等）
· 了解多智能体系统、工具使用和函数调用
· 有为非确定性AI输出设计评估和测试策略的经验
· 熟悉API设计、异步处理和可扩展后端架构
· 能够在快速变化、模糊不清、早期阶段的环境中工作
加分项
· 有实时流媒体（WebSockets、SSE）用于对话界面的经验
· 有微调、模型压缩或推理优化的经验
· 熟悉云基础设施（AWS/GCP/Azure）和CI/CD
· 有开源AI项目的贡献经历
查看英文原文
About the Role
We're looking for an AI Engineer to design, build, and scale the systems that power our AI-driven product. You'll work at the intersection of large language models, multi-agent systems, and production software, turning cutting-edge AI capabilities into reliable, user-facing experiences.
What You'll Do
· Design and implement LLM-powered pipelines, including prompt engineering, context management, and response synthesis
· Build and optimize multi-agent orchestration systems where AI components interact, reason, and produce coherent outputs
· Develop and maintain integrations with foundation model APIs (Anthropic, OpenAI, and others), managing latency, cost, and reliability at scale
· Implement retrieval-augmented generation (RAG) and memory systems for persistent, context-aware behavior
· Partner with the Evaluation/QA Engineer to build frameworks that measure output quality, coherence, and factual grounding
· Collaborate with product and design to translate workflows into robust technical experiences
· Ship production features across the stack, from the model layer to the application
Monitor, debug, and improve system performance, token efficiency, and guardrails
- 3+ years of software engineering experience, with hands-on work building LLM-powered applications
- Strong proficiency in Python and/or TypeScript/JavaScript
- Experience with LLM APIs, prompt engineering, and agentic frameworks (e.g., LangChain, LlamaIndex, or custom orchestration)
- Familiarity with vector databases and RAG architectures (Pinecone, Weaviate, pgvector, etc.)
- Understanding of multi-agent systems, tool use, and function calling
- Experience designing evaluation and testing strategies for non-deterministic AI outputs
- Solid grasp of API design, async processing, and scalable backend architecture
- Comfort working in a fast-moving, ambiguous, early-stage environment
- Experience with real-time streaming (WebSockets, SSE) for conversational interfaces
- Background in fine-tuning, model distillation, or inference optimization
- Familiarity with cloud infrastructure (AWS/GCP/Azure) and CI/CD
- Contributions to open-source AI projects