大模型应用开发工程师LLM Application Engineer

我们正在寻找一名大模型应用开发工程师,负责大模型的训练及优化。 We are looking for an LLM Application Engineer  responsible for the training and optimization of large models. This is a remote, part -time position. 职位类型 办公远程 / 弹性 项目周期预估2月(可拓展长期合作) 合作方式为项目制结算(按阶段 / 里程碑支付) Work Mode *   Remote / Flexible Hours * weeks (extendable to long -term collaboration) *   Project -based settlement (paid by phase/milestone) 岗位职责: 1.设计并实现批量 HTML 文件语义解析Pipeline。 2.集成大语言模型(如 GPT / Claude / Gemini / Ollama 等)结合规则或正则抽取结构化字段。 3.结构化存储解析结果(JSON、CSV 或数据库)。 4.动态优化大模型的提示词工程模板以提升解析准确率。 5.保障稳定性,实现并行解析架构与错误重试机制(Python/TypeScript)。 6.输出技术文档和可复用的脚本模块。 7.确保数据处理/调优/部署等技术对接符合需求。 Responsibilities *   Design and implement a batch HTML file semantic parsing pipeline. *   Utilize large language models (e.g., GPT, Claude, Gemini, Ollama) combined with rules or regular expressions to extract structured fields. *   Store the parsed results in structured formats (JSON, CSV, or databases). *   Optimize prompt engineering for large models to improve parsing accuracy. *   Integrate with Python or TypeScript scripts to achieve parallel parsing and error retry mechanisms. *   Produce technical documentation and reusable script modules. *   Validate full -process solutions from data sampling to batch outputRequirements 任职要求: 1.硕士及以上学历,计算机科学/人工智能相关专业;具备3年以上大模型开发经验。 2.熟练掌握 Python 或 TypeScript,能独立编写爬取/解析/调用 LLM 脚本。 3.熟悉 HTML DOM 结构、CSS Selector、XPath。 4.了解 OpenAI API / Claude API / LangChain / LlamaIndex 等框架。 5.有较强的数据清洗与文本处理经验(如 BeautifulSoup、Playwright、Cheerio、正则等)。 6.能独立完成从数据样本到模型提示调优、批量输出的完整流程。 7.有稳定的网络环境与良好的代码规范意识。 8.具å¤å¼ºé€»è¾'思维与跨团队协作能力,能快速理解业务需求并输出技术方案。 9.具备良好英语书面水平(如雅思5.5/英语六级/其他评级同等分数),无障碍浏览海外网站或邮件等信息。 加分项: 1.有大规模爬取 / 文档解析项目经验。 2.熟悉 Cloudflare Worker、Vercel、或 AWS Lambda 等无服务器架构。 3.对 AI agent、RAG、或信息抽取类产品有兴趣或经验。 4.具备大模型预训练、微调或分布式训练经验者优先。 5.在NeurIPS/ICML/ACL等顶会发表过论文,或参与过开源项目(如Hugging Face生态)者优先。 Requirements Must have: *   Master's degree or higher in Computer Science/Artificial Intelligence or a related field; with at least 3 years of experience in large model development. *   Proficient in Python or TypeScript, capable of independently writing scripts for crawling, parsing, and calling LLMs. *   Familiar with HTML DOM structure, CSS Selector, and XPath. *   Knowledge of frameworks such as OpenAI API, Claude API, LangChain, or LlamaIndex. *   Strong experience in data cleaning and text processing (e.g., BeautifulSoup, Playwright, Cheerio, regex, etc.). *   Ability to independently complete the entire workflow from data sampling to model prompt tuning and batch output. *   Stable network environment and a strong sense of code standardization. *   Strong logical thinking and cross -team collaboration skills, able to quickly understand business requirements and output technical solutions. *   Good written English skills (such as IELTS 5.5 or above/CET -6/equivalent scores in other rating systems), with the ability to browse overseas websites/emails and other information without obstacles. Bonus points: *   Experience with large -scale crawling / document parsing projects. *   Familiarity with serverless architectures like Cloudflare Worker, Vercel, or AWS Lambda. *   Interest or experience in AI agent, RAG, or information extraction products. *   Experience in large model pre -training, fine -tuning, or distributed training is a plus. *   Having published papers in top conferences such as NeurIPS/ICML/ACL, or participation in open -source projects (e.g., Hugging Face ecosystem) is preferred.Benefits 薪酬福利: 1.依据评估与交付范围协商;若交付质量出色,可长期合作并共同参与产品发布迭代。 2.项目制结算(按阶段/里程碑支付)+ 弹性工作时间 + 远程工作。 Compensation and Benefits *   Negotiable based on project scope and evaluation; potential for long -term collaboration and participation in product release iterations for high -quality deliveries. *   Project -based settlement (paid by phase/milestone) + flexible working hours + remote work.