资深通用 AI 工程师
Senior Agentic AI Engineer
这是一个远程职位。高级 AI 和代理系统工程师经验：5+年 位置：古尔加昂，印度 / 全程远程团队：AI 平台 /工程有关角色 我们正在寻找一个高级 AI 和机构系统工程师来设计和建造生产级人工智能代理人，可以推理，使用工具，与 API 互动，并执行复杂的多步骤工作流程。你将在 LLM 工程，代理架构，后端系统和云基础设施交叉点上工作，具有重要的建筑，可靠性，评估和扩展性的所有权。要求关键责任 · 设计和构建能够使用工具，网络/API 交互，内存和多步骤任务执行的自主 AI 代理。· 建筑工程工作流程包括计划者/执行者/验证器，多代工，HITL,重试和失败恢复模式。· 将 LLM 提供商如 OpenAI,Anthropic 和 Gemini 集成起来，包括结构化输出，函数调用，文本管理和快速优化。· 建立 BildRAG 和知识管道，采用嵌入式处理，矢量数据库，混合搜索，重新排名和内存代理。· 使用 Python 和/或 Node.js/TypeScript,REST/GraphQL API,异步处理和背景工作开发可扩展后台服务。· 设计可靠的分布式系统，以适合状态持久性，缓存，无机率，事件驱动的工作流程和故障耐受性模式。· 为人工智能系统构建可观测和评估框架，包括追踪，行为回归测试，延迟，质量和代币成本监控。· 在开发，阶段化和生产环境中进行自有 CI/CD 和云部署。· 评估 MCP,A2A,推理模型和代理框架等新兴技术并建议实用采用。所需技能 · 5年以上软件/后台工程经验，具有强大的系统设计专业知识。· 经验实践的人工智能代理商，超越基本聊天机器人实现。· 至少有一家主要 LLM 提供商 SDK:OpenAI,Anthropic 或 Gemini. · 对快速工程，工具/函数调用，结构化输出，文本管理和代币优化有强大的了解。· 与 LangGraph,CrewAI,AutoGen,ReAct 或计划执行等代理框架/模式的经验。· StrongPython 和/或 Node.js/TypeScript 开发技能。·与 RAG,嵌入式，向量数据库，混合检索和重新排名的经验
查看英文原文
This is a remote position.
Senior AI & Agentic System Engineer
Experience:5+ Years
Location:Gurgaon, India / Fully Remote
Team:AI Platform / EngineeringAbout the Role
We are looking for aSenior AI & Agentic System Engineerto design and build production-grade AI agents that can reason, use tools, interact with APIs, and execute complex multi-step workflows.
You’ll work at the intersection ofLLM engineering, agent architecture, backend systems, and cloud infrastructure, with significant ownership over architecture, reliability, evaluation, and scalability.
Requirements
Key Responsibilities
· Design and build autonomous AI agents capable of tool use, web/API interaction, memory, and multi-step task execution.
- Architect agentic workflows includingplanner/executor/verifier, multi-agent, HITL, retry, and failure-recovery patterns.
- Integrate LLM providers such asOpenAI, Anthropic, and Gemini, including structured outputs, function calling, context management, and prompt optimization.
- BuildRAG and knowledge pipelinesusing embeddings, vector databases, hybrid search, reranking, and agent memory.
- Develop scalable backend services usingPython and/or Node.js/TypeScript, REST/GraphQL APIs, async processing, and background jobs.
- Design reliable distributed systems with appropriate patterns for state persistence, caching, idempotency, event-driven workflows, and fault tolerance.
- Build observability and evaluation frameworks for AI systems, including tracing, behavioral regression testing, latency, quality, and token-cost monitoring.
- Own CI/CD and cloud deployments across development, staging, and production environments.
- Evaluate emerging technologies such asMCP, A2A, reasoning models, and agent frameworksand recommend practical adoption.
Required Skills
· 5+ years of software/backend engineering experience with strong system-design expertise.
- Hands-on experience buildingproduction AI agents, beyond basic chatbot implementations.
- Strong experience with at least one major LLM provider SDK:OpenAI, Anthropic, or Gemini.
- Strong understanding ofprompt engineering, tool/function calling, structured outputs, context management, and token optimization.
- Experience with agent frameworks/patterns such asLangGraph, CrewAI, AutoGen, ReAct, or plan-and-execute.
- StrongPython and/or Node.js/TypeScriptdevelopment skills.
- Experience withRAG, embeddings, vector databases, hybrid retrieval, and reranking.
- Strong knowledge ofREST/GraphQL APIs, asynchronous processing, databases, and distributed systems.
- Hands-on AWS experience, particularlyLambda, ECS/Fargate, API Gateway, S3, SQS, SNS/EventBridge, IAM, and CloudWatch.
- Experience withDocker, GitHub Actions, and Terraform.
- Understanding of AI security, includingprompt injection, guardrails, output validation, and safe tool execution.
AI/LLMOps & Evaluation
Experience with tools such as:
- Langfuse, LangSmith, Helicone, or Arize Phoenix
- RAGAS, DeepEval, Promptfoo, or Braintrust
- LLM tracing, behavioral evaluations, regression testing, and cost monitoring.
Nice to Have
· Browser automation usingPlaywright or Puppeteer.
- Experience with multimodal/vision-based agents.
- Knowledge ofGraphRAG and knowledge graphs.
- Experience withTemporal or durable workflow execution.
- Open-source contributions to AI tooling.
- Background in ML/data science alongside software engineering.
Tech Stack
LLMs:OpenAI, Anthropic Claude
Agent Infrastructure:Custom orchestration, MCP
Backend:Python, Node.js/TypeScript
Cloud:AWS Lambda, ECS/Fargate
Data:PostgreSQL, S3, SQS, DynamoDB
CI/CD:GitHub Actions
Observability:Datadog, CloudWatch, Langfuse
IaC:Terraform
Browser Automation:Playwright
Originally posted on Himalayas