# George Zhang

企业级 Agent 平台架构负责人

10 年大型平台与 AI 工程经验，历经阿里、商汤，现负责企业级 Agent 平台架构、产品化与团队交付。

Location: 中国
Experience: 10 years

## Primary focus
- 企业级 Agent 平台架构负责人: Agent Runtime, Agentic RAG, Tool-use / MCP, Permission & Audit, Evaluation & LLMOps, Enterprise Integration

## Featured work
- 生产级 Agent 平台架构: 负责跨行业 AI Agent 客服 SaaS 平台从架构演进到生产运营，并推动团队 AI 开发 Harness 覆盖 40 个仓库。 (/zh/work/enterprise-agent-platform)
- AI 数据标注与模型服务平台: 自主创业阶段从 0 到 1 交付多模态数据标注与模型服务平台，打通训练数据、模型产物、自动标注、质量校验和 Serverless 推理链路。 (/zh/work/ai-data-model-platform)
- AIGC SaaS 与模型服务平台: 负责生成式 AI SaaS 的 Web 工程、微服务拆分、模型服务调用链和平台化交付，参与产品从 0 到 1 建设。 (/zh/work/aigc-saas-platform)
- 阿里高流量商业化与直播工程: 负责阿里淘宝与本地生活高流量跨端 Web、微前端、广告归因、日志 SDK、JsBridge 和性能优化。 (/zh/work/commerce-platform-engineering)

## Lab
- DPO for Tool-use Preference: 将工具调用问题样本转化为偏好数据，连接回放评测、策略优化与模型改进。 (/zh/lab/dpo-tool-use)
- Agent 回放评测框架: 用生产 trace 构建可复现的 Agent 回归评测，支撑 prompt、工具和模型变更。 (/zh/lab/eval-harness)

## Writing
- 企业级 Agent 平台能力地图: 面向生产级 Agent 平台的 runtime、工具治理、评测回放、团队工程和业务价值框架。 (/zh/writing/agent-platform-positioning-assessment)
- 企业 Agent 评测手册: 把 Agent 评测从主观试用变成可回放、可分层、可发布门禁的工程系统。 (/zh/writing/agent-evaluation-playbook)

## Contact
Email: sunionpeaks@gmail.com

---

# George Zhang

Enterprise Agent Platform Architect

10 years across large-scale platforms and AI engineering at Alibaba and SenseTime, now leading enterprise Agent platform architecture, productization, and delivery.

Location: China
Experience: 10 years

## Primary focus
- Enterprise Agent Platform Architect: Agent Runtime, Agentic RAG, Tool-use / MCP, Permission & Audit, Evaluation & LLMOps, Enterprise Integration

## Featured work
- Production-grade Agent Platform Architecture: Owned a cross-industry AI Agent customer-service SaaS platform from architecture evolution through production operations and expanded the team AI development harness across 40 repositories. (/en/work/enterprise-agent-platform)
- AI Data Labeling and Model-serving Platform: Built and delivered a multimodal data-labeling and model-serving platform from 0 to 1, connecting training data, model artifacts, auto-labeling, quality validation, and serverless inference. (/en/work/ai-data-model-platform)
- AIGC SaaS and Model-serving Platform: Owned web engineering, microservice decomposition, model-serving call chains, and platform operations for 0-to-1 generative AI SaaS delivery. (/en/work/aigc-saas-platform)
- Alibaba High-traffic Commerce and Live Engineering: Owned high-traffic cross-platform web engineering, micro-frontends, ad attribution, logging SDKs, JsBridge, and performance optimization across Alibaba Taobao and Local Services. (/en/work/commerce-platform-engineering)

## Lab
- DPO for Tool-use Preference: Turns tool-use issue cases into preference data connecting replay evaluation, policy optimization, and model improvement. (/en/lab/dpo-tool-use)
- Agent Replay Evaluation Harness: Built reproducible Agent regression evaluation from production traces for prompt, tool, and model changes. (/en/lab/eval-harness)

## Writing
- Enterprise Agent Platform Capability Map: A production Agent platform framework across runtime, tool governance, replay evaluation, team engineering, and business value. (/en/writing/agent-platform-positioning-assessment)
- Enterprise Agent Evaluation Playbook: Turning Agent evaluation from subjective trial into replayable, layered, release-gating engineering. (/en/writing/agent-evaluation-playbook)

## Contact
Email: sunionpeaks@gmail.com