[Hiring] Research Intern - Multimodal Data Retrival (Alibaba Qwen Team)

The Mission Data quality determines the upper bound of model potential. We are building a high-performance Multimodal Semantic Retrieval System designed to diagnose, curate, and filter 100 Billion+ multimodal data points with extreme speed and flexibility.

The Challenge We are looking for collaborators to build the “search engine” for our training data:

• Scale: Architecting systems to handle 100B+ multimodal data entries.
• Deep Analysis: Enabling rapid quantification of patterns, bias detection, and live traffic analysis.
• Retrieval: Developing efficient semantic retrieval algorithms to pinpoint specific high-value data within massive datasets.

Who We Are Looking For We are primarily seeking Research Interns (exceptional Full-time candidates will also be considered):

• Background: Strong foundation in Information Retrieval (IR), NLP, Computer Vision, or Big Data Systems.
• Expertise: Experience with Vector Databases, Multimodal Representation Learning (CLIP, SigLIP), or LLM Supervised Fine-Tuning (SFT) pipelines.
• Mindset: Focused on data-centric AI and infrastructure optimization.

Resources & Benefits

• Top-tier Infrastructure: Access to massive computing resources and state-of-the-art AI tools (Cursor, Gemini/Claude Pro, etc.).
• Competitive Package: Market-leading salary, top-spec MacBook, and full catering support.

Apply

• Email: zhengjie.hsj@alibaba-inc.com
• Note: Please attach your resume and highlight relevant experience in data systems or model fine-tuning.