{
    "schema_version": "via_ai_product_manager_jobs_v1",
    "data_type": "ai_product_manager_jobs",
    "job": {
        "job_title": "AI engineer",
        "job_level": "高级",AI生成
        "work_category": "全职",AI生成
        "department": "未披露",
        "location": "天津·滨海新区·经济开发区",
        "work_mode": "现场办公",AI生成
        "job_category": "AI应用软件工程师"AI生成
    },
    "company": {
        "company_name": "凯捷",
        "company_industry": "计算机软件",
        "company_size": "10000人以上",
        "company_website": "未披露",
        "company_description": "凯捷(Capgemini)是一家由AI驱动的全球业务与技术转型合作伙伴，旨在创造切实的商业价值。我们为企业构想未来，并通过AI、技术和人才将其变为现实。凯捷是一个负责任的多元化组织，集团成立于1967年，总部位于法国巴黎，在50多个国家拥有近42万名团队成员。我们凭借深厚的行业专业知识和强大的合作伙伴生态系统，提供端到端的服务和解决方案，充分发挥战略、技术、设计、工程及业务运营等方面的能力。集团2024年全球收入达到221亿欧元。"
    },
    "compensation": {
        "salary_range": {
            "min": 20000,
            "max": 30000,
            "currency": "元",
            "cycle": "月"
        },
        "benefits": "通讯补贴 | 节日福利 | 零食下午茶 | 年终奖 | 带薪年假 | 带薪病假 | 定期体检 | 补充医疗保险 | 五险一金 | 生日福利 | 体系技培训"
    },
    "requirements": {
        "working_years": 5,
        "degree": "本科",
        "language": "英文",AI生成
        "visa_support": "未披露",
        "responsibilities": "LLM\nAI\nNLP\nagent\nRole Overview\nAs an AI Application Engineer, you’ll be responsible for integrating, fine-tuning, and operationalizing large language models and other AI components into production-grade software systems. This role focuses on post-training workflows—such as fine-tuning, prompt engineering, retrieval-augmented generation (RAG), vector database integration, and model deployment—rather than foundational model development.\nYou’ll work closely with backend engineers, product teams, and data teams to build intelligent apps that are fast, accurate, consistent and production-ready.\nKey Responsibilities\n- Integrate LLMs and other AI components into web and backend applications.\n- Fine-tune open models using techniques like LoRA, QLoRA, or supervised fine-tuning.\n- Build RAG pipelines using vector databases.\n- Design and optimize prompts for various tasks (prompt chaining, templating, few-shot setups).\n- Package and deploy models for inference\n- Evaluate and monitor AI model performance in live environments.\n- Ensure scalable, secure, and compliant use of AI systems in production.\n- Stay informed on LLM tooling ecosystems (LangChain, LlamaIndex, OpenAI APIs, Hugging Face, Spring AI etc.)\nRequired Qualifications\n- Bachelor’s or Master’s in Computer Science, Engineering, or a related field.\n- 2+ years of experience building AI-augmented applications or services.\n- Proficient in Python and familiar with frameworks like LangChain, Transformers, or Hugging Face datasets.\n- Understanding of vector embeddings and experience with vector search tools.\n- Strong grasp of model deployment workflows (e.g., Docker, REST APIs, cloud services like AWS/Microsoft Azure).\n- Familiarity with prompt engineering strategies and LLM behavior tuning.\n- Experience integrating with APIs from OpenAI, Anthropic, or open-source models like Mistral or LLaMA.\nPreferred Qualifications\n  - Experience fine-tuning models using PEFT methods (LoRA/QLoRA).\n- Familiarity with LLMOps, observability, and model evaluation\n- Exposure to security, privacy, and responsible AI practices.\n- Prior experience building chatbots, copilots, document analyzers, or other AI apps.\n- Contributions to open-source AI tools or GenAI workflows.",
        "requirements_text": "- Bachelor’s or Master’s in Computer Science, Engineering, or a related field.\n- 2+ years of experience building AI-augmented applications or services.\n- Proficient in Python and familiar with frameworks like LangChain, Transformers, or Hugging Face datasets.\n- Understanding of vector embeddings and experience with vector search tools.\n- Strong grasp of model deployment workflows (e.g., Docker, REST APIs, cloud services like AWS/Microsoft Azure).\n- Familiarity with prompt engineering strategies and LLM behavior tuning.\n- Experience integrating with APIs from OpenAI, Anthropic, or open-source models like Mistral or LLaMA.\nPreferred Qualifications\n  - Experience fine-tuning models using PEFT methods (LoRA/QLoRA).\n- Familiarity with LLMOps, observability, and model evaluation\n- Exposure to security, privacy, and responsible AI practices.\n- Prior experience building chatbots, copilots, document analyzers, or other AI apps.\n- Contributions to open-source AI tools or GenAI workflows.",
        "skills": [AI生成
            "Agent系统构建",
            "Prompt工程",
            "RAG与检索增强",
            "Workflow与任务编排",
            "云平台与云计算",
            "企业服务/SaaS知识",
            "前后端/全栈开发",
            "大模型原理与能力边界",
            "模型/Agent评测体系",
            "模型微调与对齐",
            "编程与代码实现",
            "部署上线与环境配置"
        ],
        "preferred_qualifications": "Preferred Qualifications",
        "job_highlights": "5-10年 | 本科"
    },
    "contact": {
        "recruiter_name": "张女士",
        "contact_methods": [
            {
                "type": "apply_link",
                "value": "https://www.zhipin.com/job_detail/44483772c484eae40nN53dy5FlpS.html"
            }
        ]
    },
    "posting": {
        "posting_id": "44483772c484eae40nN53dy5FlpS",
        "post_date": "",
        "hire_time": "未披露",
        "update_date": "20260915",
        "job_url": "https://www.zhipin.com/job_detail/44483772c484eae40nN53dy5FlpS.html"
    },
    "fulltext": "LLM\nAI\nNLP\nagent\nRole Overview\nAs an AI Application Engineer, you’ll be responsible for integrating, fine-tuning, and operationalizing large language models and other AI components into production-grade software systems. This role focuses on post-training workflows—such as fine-tuning, prompt engineering, retrieval-augmented generation (RAG), vector database integration, and model deployment—rather than foundational model development.\nYou’ll work closely with backend engineers, product teams, and data teams to build intelligent apps that are fast, accurate, consistent and production-ready.\nKey Responsibilities\n- Integrate LLMs and other AI components into web and backend applications.\n- Fine-tune open models using techniques like LoRA, QLoRA, or supervised fine-tuning.\n- Build RAG pipelines using vector databases.\n- Design and optimize prompts for various tasks (prompt chaining, templating, few-shot setups).\n- Package and deploy models for inference\n- Evaluate and monitor AI model performance in live environments.\n- Ensure scalable, secure, and compliant use of AI systems in production.\n- Stay informed on LLM tooling ecosystems (LangChain, LlamaIndex, OpenAI APIs, Hugging Face, Spring AI etc.)\nRequired Qualifications\n- Bachelor’s or Master’s in Computer Science, Engineering, or a related field.\n- 2+ years of experience building AI-augmented applications or services.\n- Proficient in Python and familiar with frameworks like LangChain, Transformers, or Hugging Face datasets.\n- Understanding of vector embeddings and experience with vector search tools.\n- Strong grasp of model deployment workflows (e.g., Docker, REST APIs, cloud services like AWS/Microsoft Azure).\n- Familiarity with prompt engineering strategies and LLM behavior tuning.\n- Experience integrating with APIs from OpenAI, Anthropic, or open-source models like Mistral or LLaMA.\nPreferred Qualifications\n  - Experience fine-tuning models using PEFT methods (LoRA/QLoRA).\n- Familiarity with LLMOps, observability, and model evaluation\n- Exposure to security, privacy, and responsible AI practices.\n- Prior experience building chatbots, copilots, document analyzers, or other AI apps.\n- Contributions to open-source AI tools or GenAI workflows."
}