Foundation Model Evaluation & Adaptation Tech Lead

Responsibilities : - Overall architecture design bridging Agent and self-developed foundation models - Lead the Harness layer architecture design and evaluation system setup - Encapsulate foundation model inference interfaces, sampling strategies, and context management to provide a unified invocation layer for upper-level Agents - Collaborate with the foundation model team to standardize model capability evaluation Requirements : - Experience in model evaluation / benchmark testing / Harness architecture design - Familiarity with LLM evaluation methodologies - Technical team management experience