大模型算法工程师（可靠性方向）-小荷健康
Core
Define real-world capability frameworks for medical AI models handling multimodal, multi-turn interactions and continuous health management tasks.
Role type
Senior Large Language Model Algorithm Engineer (Reliability & Medical AI)
Builds
Dynamic medical knowledge and evidence systems, multimodal LLMs, Agents, and LLM Harnesses for clinical decision support.
Domain
Healthcare / Artificial Intelligence / Large Language Models
Deliverable
production ML models
Required skills
Large Language Model (LLM) architecture, Agent design, RAG (Retrieval-Augmented Generation), Tool Use, Memory management, Multimodal understanding, Model alignment, AI reliability evaluation, Real-world data analysis, Medical knowledge engineering
Preferred skills
Clinical evidence grading (GRADE), Systematic Review methodology, Hallucination detection, Red Teaming, Fact-checking, Medical/AI conference publications (ICML, ICLR, NeurIPS, ACL)
Technologies
LLM Harness, Multimodal models, Agent frameworks
Responsibilities
Define capability frameworks for evidence usage, dialogue decision-making, and complex task completion; Construct updatable and verifiable medical knowledge bases; Identify model failure modes from user interactions to drive algorithmic iteration; Embed model capabilities and feedback loops into production systems.
Seniority
Senior, hands-on IC
