WeTalk-大模型应用研发工程师-（Agent Harness）
Core
Design and develop the WeTalk Agent Harness, including Agent Loop, orchestration, runtime environments, and evaluation systems to optimize LLM applications.
Role type
Senior IC machine-learning engineer (LLM application & Agent systems)
Builds
Stable, general-purpose engineering capabilities for WeTalk AI products
Domain
Artificial Intelligence / Large Language Models / Agent Systems
Deliverable
production ML models
Required skills
Go, Python, Node.js, MySQL, Redis, Pulsar, LLM, Agent, Tool Use, MCP, Memory management
Preferred skills
Context Engineering, long-task state management, prompt optimization
Responsibilities
Design and develop core Agent Harness capabilities including loops and orchestration; Build Context Engineering features like compression and memory; Establish evaluation systems using traces and user feedback; Explore and implement WeTalk AI capabilities.
