Focus on core business scenarios including material design, battery R&D, energy storage systems and intelligent manufacturing. Adopt new AI technologies and machine learning algorithms to solve practical industrial problems, and promote the full-integration of AI with renewable energy scientific research, production and simulation, technology implementation, scenario empowerment and efficiency improvement.
聚焦新能源材料、电池研发、储能系统、智能制造等核心业务场景，依托AI算法、大模型、机器学习技术解决产业实际问题，推动AI与新能源科研、生产、仿真全流程融合落地，主打技术落地、场景赋能、效率提效。
Responsibilities | 职责：
The responsibilities listed cover the full scope of this role. Applicants only need to independently undertake one or two core items, with other tasks completed via teamwork.
1. AI-driven Computations: Based on physical view of related material systems of new energy scenarios, and starting from fundamental thermodynamics and kinetics phenomenological equations to complete analysis and simulation; use AI technology to accelerate multiscale materials simulations; complete high-quality physics data generation and close-loop; complete interdisciplinary collaborative verification; complete technical tracking and implementation with related R&D team.
2. AI-enabled R&D: Dig deep into core new energy scenarios. Design AI algorithm solutions targeting pain points such as novel material disovery & optimization, synthetic pathway prediction & optimization, process optimization, battery life prediction; complete model development, training, tuning and iterative implementation.
3. Integration with AI for Science: Combine AI4S technologies to connect multi-scale simulation, crystal simulation, electrochemical experiments and other scientific research work. Accelerate the R&D iteration of new materials and new battery systems via machine learning, deep learning and generative AI.
4. Data Mining & Modeling: Conduct data mining and feature engineering based on time-series data, experimental data and simulation data; build scenario-oriented prediction, classification and optimization models, and deliver implementable algorithm results, patents and technical reports.
5. Cutting-edge Technology Exploration: Keep track of cutting-edge applied AI technologies. Explore innovative application scenarios of large models, agents, reinforcement learning and digital twins in the new energy sector, and complete technical verification and project promotion with the R&D team.
6. Cross-team Adaptation & Implementation: Cooperate with simulation, experimental and engineering teams to conduct adaptive debugging of algorithms, ensuring stable operation of models in scientific trials and small-scale pilots.
职责说明：下述职责为该方向完整业务范围，应聘人员只需深耕并独立承担其中1~2项核心工作，其余工作以团队协作方式共同推进。
1. AI驱动计算：立足新能源材料体系物理图像，及基本热力学和动力学方程，结合AI技术加速多尺度材料模拟，完成高质量物理数据的生成与闭环，完成跨学科协同验证，并配合相关研发团队完成技术追踪与落地
2. 研发AI赋能：深耕新能源核心场景，针对新型材料设计、合成路径预测与优化、工艺优化、电池寿命预测等业务痛点，设计AI算法解决方案，完成模型研发、训练、调优与迭代落地。
3. 科学智能融合：结合AI4S技术，对接多尺度仿真、电化学实验等科研工作，通过机器学习、深度学习、生成式AI加速新材料、新体系电池的研发迭代。
4. 数据建模挖掘：基于时序数据、实验数据、仿真数据开展数据挖掘、特征工程，搭建场景化预测、分类、优化模型，输出可落地的算法成果、专利及技术报告。
5. 前沿技术探索：跟踪行业前沿AI应用技术，探索大模型、智能体、强化学习、数字孪生在新能源领域的创新落地场景，配合研发团队完成技术验证与项目推进。
6. 跨团队落地适配：协同仿真、实验、工程落地团队，完成算法模型的适配调试，保障模型在科研测试、小规模试点中的稳定运行。
Qualifications | 任职要求：
As for the two core proficiency channels listed below (Computational Physics Proficiency & AI Algorithm Proficiency), applicants only need to independently undertake one core item.
1. Educational & Academic Background: Master’s degree or above, majoring in Artificial Intelligence, Computer Science, Big Data, Applied Mathematics, Condensed Matter Physics, Computational Physics, Computational Chemistry, Materials Science, New Energy, Automation or related disciplines，possess in-depth understanding of solid-state physics, electrochemical thermodynamics and kinetics. Candidates with practical experience in AI for Science and new energy algorithm implementation are preferred.
2. Computational Physics Proficiency: Solid command of principles and application boundaries of density functional theory (DFT) and molecular dynamics simulation (MD), proficient in at least one mainstream DFT software (VASP、Quantum ESPRESSO、CP2K, etc) and MD software (LAMMPS、OPENMM、GPUMD, etc)；skilled in application of machine learning in scientific computations (e.g., machine learning force field) and basic PyTorch; basic knowledge of advanced sampling (enhanced sampling, metadynamics, MC-MD) or coarse-grained modeling；
3. AI Algorithm Proficiency: Proficient in Python and mainstream deep learning frameworks (PyTorch/JAX/TensorFlow), with a solid understanding of core machine learning and time-series forecasting algorithms; possess deep R&D experience in advanced architectures such as Graph Neural Networks (GNNs), Equivariant Neural Networks, Transformers, and Diffusion Models, along with a clear understanding of physical constraints (e.g., energy conservation, symmetry) and the applicability boundaries of computational physics methods.
4. Practical Modeling Capability: Capable of independent algorithm modeling, model training, parameter tuning and effect iteration. Prior experience in AI projects for battery R&D, energy storage optimization, material design or industrial fault detection is preferred.
5. Data Analysis Competence: Solid data analysis capability to process scientific simulation data and industrial time-series data; proficient in the full workflow of data cleansing, feature construction and model verification.
6. Collaboration & Problem-solving: Clear logical thinking and strong cross-team collaboration skills. Able to accurately interpret scientific research requirements and deeply integrate AI technologies with new energy R&D scenarios; equipped with robust problem-solving abilities.
对于下述两条核心专业通道（计算物理能力&AI算法能力），符合任意一条核心要求即可：
1. 学历专业背景：硕士及以上学历，人工智能、计算机、大数据、应用数学、凝聚态物理、计算物理、计算化学、材料科学、新能源、自动化等相关专业，对固体物理，电化学热力学与动力学等有深刻理解，有AI for Science、新能源算法落地经验者优先。
2. 计算物理能力：掌握密度泛函理论（DFT）和分子动力学模拟（MD）的底层原理和应用边界，精通至少一种主流DFT软件（VASP、Quantum ESPRESSO、CP2K等）和MD软件（LAMMPS、OPENMM、GPUMD等）；熟悉机器学习在科学计算中的应用（如MLFF等）及基础PyTorch；对高级采样（增强采样、元动力学、MC-MD）或粗粒化（Coarse-graining）建模有一定了解；
3. AI算法能力：精通 Python 编程，熟练掌握 PyTorch / JAX / TensorFlow 等主流深度学习框架，扎实理解机器学习与时序预测等核心算法原理；在图神经网络（GNN）、等变神经网络、Transformer、扩散模型等前沿架构上有深入研发经验，且对物理约束（如能量守恒、对称性）及计算物理方法的适用边界有清晰认知。
4. 建模实战经验：具备独立算法建模、模型训练、参数调优、效果迭代能力，有电池研发、储能优化、材料设计、工业故障检测相关AI项目落地经验者优先。
5. 数据分析功底：具备良好的数据分析能力，能够处理科研仿真数据、工业时序数据，熟悉数据清洗、特征构建、模型验证全流程。
6. 协同解决问题：逻辑清晰、擅长跨团队协同，能够精准理解科研业务需求，将AI技术与新能源研发场景深度结合，具备较强的问题解决能力。
Preferred Qualifications | 优先条件：
1. Practical experience in large model fine-tuning, industrial agent development and reinforcement learning-based dispatching optimization;
2. Familiarity with multi-scale simulation, lithium/perovskite battery material development and electrochemical mechanisms;
3. Published papers on AI+Energy or AI+Materials, or possessing relevant technical patents.
1. 大模型开发经验：有大模型微调、行业智能体开发、强化学习调度优化实战经验；
2. 行业专业认知：熟悉多尺度模拟、钙钛矿/锂电材料研发、电化学机理相关知识；
3. 成果专利产出：发表过AI+能源、AI+材料相关论文，或拥有相关技术专利。
