理想汽车Multimodal AI Agent Algorithm Engineer

Team Introduction： Through algorithm optimization and collaboration with business teams, the team conducts comprehensive quality and ecosystem governance for ByteDance's e-commerce products. This involves combating risks, violations, and low-quality issues, as well as constructing and optimizing a healthy e-commerce ecosystem. The team aims to maximize platform governance effectiveness while improving operational efficiency and reducing costs. Additionally, the team is dedicated to advancing cutting-edge AI technologies to drive business transformation and development through technical innovation, covering diverse fields including but not limited to NLP, CV, multimodal models, large models, graph algorithms, and sequence algorithms. 团队介绍： 平台治理算法团队，通过优化算法，和业务团队协作，对字节旗下的电商产品进行全方位的质量和生态的治理，既包括风险、违规和低质问题的打击，也包括健康电商生态的建设和优化，在最大程度的优化平台治理的效果的同时提升治理的工作效率，降低成本。另外一方面，平台治理算法团队致力于攻坚前沿的AI技术，以技术驱动推动业务的变革和发展，领域涉及广泛，包括但不限于NLP/CV/多模态/大模型/图算法/序列算法等。 课题目标和必要性： 电商智能审核业务比较复杂，随着审核技术的不断演进，各个领域面临着新的风险问题和对抗形式，这对大模型的应用提出了新的挑战。例如，在电商审核业务中，涉及审核PBR变更、长文本、长时序、多语言、少样本和AIGC生成对抗等问题时，现有的开源大模型表现往往不尽人意。因此，针对这些挑战，我们亟需研发专门针对电商智能审核的大模型，以提升其在电商治理中的有效性和适应性。特别的，针对电商业务特点，我们需要探索高质量的数据自动生成、高效的MOE Embedding、Auto-prompt生成、高质量 COT输出、大模型知识蒸馏等。此外，该模型应能够满足电商审核业务的需求，实现高准确率的自主决策和可解释性的COT生成，显著减少误判。针对动态变化的审核PBR变更，它能够通过RAG模块自动检索类似的审核案例，将复杂的审核PBR分解为简单的原子任务，自动拆分出驳回和豁免原子任务，并自动调用相应的Tools来解决这些任务，从而建立“知道拒绝并且知道为何拒绝”的业内领先智能审核系统。最终，大模型智能审核系统的审核效果需要接近或者超过人工审核，往全机审的路线上演进。 课题内容： 电商智能审核多模态大模型，主要研究点包括但不限于： 1、模态融合能力：提升文本、音频、图像、视频和直播等多模态的细粒度理解能力，实现高准确率的自主决策和可解释性的COT生成； 2、Few-Shot能力：探索电商多语言、长时序和少样本问题，增强Few-Shot和Zero-Shot能力，针对多变的业务规则具备复杂指令和Auto-prompt生成能力； 3、攻防对抗能力：研究AIGC图像视频的判别，增强审核大模型对隐晦、抽象的生成式内容的攻防对抗能力； 4、 Agent能力：具备调用RAG模块，使用Tools，和Auto-planning能力；提升大模型的动态推理和反思能力。 涉及的研究方向：大模型，多模态大模型，Few-Shot，AIGC判别，AIGC数据生成，强化学习，Agent。

• Design, fine-tune, and optimize AI models (CV, NLP, multi-modal, or hybrid) to meet real-world performance, latency, and accuracy requirements. • Develop and improve AI Agent algorithms, including reasoning, planning, decision-making, tool-use, and multi-agent collaboration strategies. • Build and maintain a Model Factory approach, enabling standardized model training, fine-tuning, evaluation, versioning, and deployment. • Conduct systematic experimentation on model architectures, hyperparameters, prompt strategies, and fine-tuning techniques. • Collaborate with data engineering teams to define data requirements for training, validation, and continuous model improvement. • Design evaluation frameworks and metrics to measure model and agent performance across accuracy, robustness, efficiency, and business impact. • Optimize inference performance for production environments, including latency, memory usage, and cost efficiency. • Research and apply state-of-the-art techniques in LLMs, foundation models, reinforcement learning, multi-agent systems, and multi-modal learning. • Translate business and product requirements into algorithmic strategies and technical designs. • Document model behavior, limitations, and improvement plans to support long-term maintainability and governance.

Why Join Us We're seeking an AI Algorithm Engineer who will lead the design and implementation of cutting-edge conversational systems for TCL products such as TV, mobile phones, smart glasses and home robots. This position is ideal for someone who is motivated to innovate, tackle challenging problems, and collaborate effectively in cross-functional teams. Key Responsibilities Lead the algorithm design and data flow architecture for specific scenarios. Adapt LLMs to specific application scenarios via continued pre-training or fine-tuning. Evaluate and monitor model performance using both quantitative and qualitative metrics; recommend the best models for specific tasks. Stay current with AIGC trends and apply cutting-edge technologies and methods to work. Solve complex problems, share knowledge, and collaborate across departments.

Why Join Us We're seeking an AI Algorithm Engineer who will lead the design and implementation of cutting-edge conversational systems for TCL products such as TV, mobile phones, smart glasses and home robots. This position is ideal for someone who is motivated to innovate, tackle challenging problems, and collaborate effectively in cross-functional teams. Key Responsibilities Lead the algorithm design and data flow architecture for specific scenarios. Adapt LLMs to specific application scenarios via continued pre-training or fine-tuning. Evaluate and monitor model performance using both quantitative and qualitative metrics; recommend the best models for specific tasks. Stay current with AIGC trends and apply cutting-edge technologies and methods to work. Solve complex problems, share knowledge, and collaborate across departments.