Senior AI Data Infrastructure Engineer

Meshy is building the next generation of 3D generative AI models. Underpinning all of this is our constantly evolving data infrastructure – from the ingestion of massive amounts of raw data to large-scale distributed processing, and the construction of high-quality datasets needed for model training. Data infrastructure is one of our core technological moats and a decisive factor in the research team's ability to iterate rapidly and continuously. We are currently recruiting a Senior Data Infrastructure Engineer in Shanghai to join the Data Infra team and work with engineers worldwide to design, build, and operate distributed data systems that support large-scale AI model training. You will be handling both regular structured data and unstructured assets such as images, 3D models, and videos – the latter typically requiring specialized processing and transformation for pre-training and fine-tuning. This is a versatile role: you will have ownership of an end-to-end data pipeline from data ingestion to transformation and training readiness; you will work closely with ML research teams to prepare various datasets for cutting-edge model training; and you will find your perfect fit in a fast-paced, adaptable, and hands-on startup environment. It's important to note that this is not a data warehousing/BI/data analysis role. The main focus of daily work is on processing pipelines for large-scale unstructured data, optimizing computing and storage efficiency, and ensuring the quality of training data, rather than report and metric modeling.

Responsibilities

Design and maintain highly scalable ETL/ELT pipelines to support the ingestion and processing of structured and massive amounts of unstructured data (images, videos, 3D/2D assets); Optimize data processing using frameworks such as Spark/Ray/Flink based on cloud object storage and data lakes, implementing data sharding, caching, and a robust monitoring and alerting mechanism; Responsible for the preprocessing (format conversion, enhancement, feature extraction, etc.) and quality verification of model pre-training data, ensuring the lineage and reproducibility of datasets; Manage environments using IaC tools such as Terraform/Kubernetes, promoting CI/CD best practices; collaborate closely across teams to quickly respond to R&D needs.

Requirements

5+ years of experience in data engineering or distributed system development; basic English communication skills, capable of conducting all-English meetings and written communication; Design and maintain highly scalable ETL/ELT pipelines to support the ingestion and processing of structured and massive amounts of unstructured data (images, videos, 3D/2D assets); Optimize data processing using frameworks such as Spark/Ray/Flink based on cloud object storage and data lakes, implementing data sharding, caching, and a robust monitoring and alerting mechanism; Responsible for the preprocessing (format conversion, enhancement, feature extraction, etc.) and quality verification of model pre-training data, ensuring the lineage and reproducibility of datasets; Manage environments using IaC tools such as Terraform/Kubernetes, promoting CI/CD best practices; fostering close cross-team collaboration and rapid response to R&D needs.

Nice to have

Familiarity with the entire training workflow for large models or machine learning (Pre-training/Fine-tuning/RL); Experience in 3D/2D asset processing (geometric transformation, texture processing) or familiarity with mainstream rendering engines (Blender/Unity/Unreal) for compositing data generation; Practical experience in Kubernetes distributed workload orchestration or GPU/HPC clusters; Extensive experience using large-scale data warehouse/lake warehouse platforms (Databricks/Snowflake/BigQuery, etc.); High-quality open-source contributions in ML Infrastructure and distributed systems, or familiarity with data security and compliance.

Benefits

Competitive salary, equity, and benefits package; Opportunity to work with a talented and passionate team at the forefront of AI and 3D technology; Flexible work environment, with options for remote and on-site work; Opportunities for fast professional growth and development; Inclusive culture that values creativity, innovation, and collaboration; Unlimited, flexible time off; Stock options available for core team members; 401(k) plan for employees; Comprehensive health, dental, and vision insurance; The latest and best office equipment.