LLM Engineer
角色摘要 KDCI 外包招聘正在寻找一个 LLM 工程师来设计，细节调整和部署解决方案大型语言模型，以解决真正的业务问题。你将在整个模型生命周期中工作从快速设计和架构到评估和生产监测与产品和工程团队合作发送可靠的人工智能功能。关键责任 · 设计，细致调整和优化大型语言模型特定使用情况 · 建立和维护评估带来衡量模型质量和安全性 · 实施 RAG （恢复增强生成）管道和快速策略 · 在生产中监控模型性能并反复改进 ·与产品和工程团队合作，集成 LLM 功能 ·跟上新兴的 LLM 研究，架构和工具要求要求 · 3年以上实践经验的大语言模型和 NLP ·强有力的理解变压器架构 （GPT,BERT,LLaMA 等）·熟练使用 PyTorch 或 TensorFlow 等 Python 和 ML 框架 ·精细调度，快速工程和超参数优化方面的经验 ·在生产环境中部署 ML/LLM 系统的记录 ·对矢量数据库和 RAG 管道的经验 ·熟悉模型定量化和推断优化 · · 基于 AI 安全或内容调节系统的背景突出地点使用人工智能，RAG 和现代 ML 框架构建并部署先进的 LLM 解决方案。最初发布在喜马拉雅山脉
查看英文原文
Role Summary
KDCI Outsourcing is looking for an LLM Engineer to design, fine-tune, and deploy large language model solutions that solve real business problems. You'll work across the model lifecycle — from prompt and architecture design to evaluation and production monitoring — collaborating with product and engineering teams to ship reliable AI features.
Key Responsibilities
- Design, fine-tune, and optimize large language models for specific use cases
- Build and maintain evaluation harnesses to measure model quality and safety
- Implement RAG (retrieval-augmented generation) pipelines and prompt strategies
- Monitor model performance in production and iterate on improvements
- Collaborate with product and engineering teams to integrate LLM features
- Stay current with emerging LLM research, architectures, and tooling
Requirements
Requirements
- 3+ years of hands-on experience with large language models and NLP
- Strong understanding of transformer architectures (GPT, BERT, LLaMA, etc.)
- Proficiency in Python and ML frameworks such as PyTorch or TensorFlow
- Experience with fine-tuning, prompt engineering, and hyperparameter optimization
- Track record of deploying ML/LLM systems in production environments
Nice-to-Haves
- Experience with vector databases and RAG pipelines
- Familiarity with model quantization and inference optimization
- Background in AI safety or content moderation systems
Highlights
Build and deploy cutting-edge LLM solutions using AI, RAG, and modern ML frameworks.
Originally posted on Himalayas