多模态评测工程师-Seed
Core
Building and iterating a multimodal evaluation system for audio/video large models (ASR, TTS, music generation, multi-modal interaction, Agents) to establish industry benchmarks and drive model optimization.
Role type
Senior IC multimodal evaluation engineer (large language models)
Builds
Automated, scalable, high-reliability evaluation systems and datasets for MLLMs and GenMedia
Domain
AI / Large Language Models / Multimodal AI
Deliverable
production ML models
Required skills
C/C++, Python, Go, debugging, AI coding tools, CLI, Skills, Agent technologies, large model evaluation, data analysis, report generation
Preferred skills
Industry-impactful large model evaluation projects, experience in audio/video/text large models
Technologies
C/C++, Python, Go, AI coding tools, CLI, Skills, Agent frameworks
Responsibilities
Design and iterate multimodal evaluation schemes and metrics; Develop automated evaluation systems to improve efficiency; Analyze model performance to identify strengths and weaknesses; Build a 'data-model-evaluation' closed loop with Agents; Track and apply frontier evaluation technologies.
Seniority
Senior, hands-on IC