Multimodal LLM Algorithm Engineer Intern (Global E-Commerce, Knowledge Graph) - 2027 Start (PhD)
Core
Build multimodal content understanding, product understanding, semantic matching, pricing intelligence, and intelligent Agent capabilities for global content-commerce scenarios.
Role type
PhD intern multimodal LLM algorithm engineer
Builds
Product catalog, identical product matching systems, pricing algorithms, multimodal semantic links, generative search/recommendation, and business-facing Agents
Domain
Global E-commerce, Knowledge Graph, Generative AI
Deliverable
production ML models
Required skills
NLP, Computer Vision, generative models (Diffusion, GANs, ControlNet), multimodal learning, large models, Agents, PyTorch, TensorFlow, model training, tuning, deployment
Preferred skills
LLM/VLM, Embedding/SID, RAG, Tool Calling, Agent Planning, distributed training, inference acceleration, TensorRT, LLM serving optimization, multimodal retrieval, pricing algorithms, AIGC
Technologies
PyTorch, TensorFlow, TensorRT
Responsibilities
Build global product catalog and content understanding systems; Develop algorithms for identical product matching and cross-lingual aggregation; Build pricing algorithms for price comparison and anomaly detection; Construct multimodal semantic links across video-product and product-product relationships; Explore next-generation generative search and recommendation; Design generative algorithms for visual creatives (background synthesis, virtual try-on); Explore business-facing Agents (price comparison, product analysis, content AIGC); Own end-to-end algorithm workflow including data construction, training, evaluation, and deployment
