Principal Product Manager, AI Agent Infrastructure

We are looking for a Product Manager who can drive the foundational infrastructure behind AI agent products. As agentic AI rapidly moves from research to production, the platforms that power agent development, evaluation, and deployment have become mission-critical. You will own the product vision, strategy, and execution for these infrastructure systems — ensuring our teams can build, test, and ship high-quality AI agents at scale.

This role is ideal for someone who has worked on AI/ML platforms, developer tools, or data infrastructure, and wants to apply that experience to the frontier of agentic AI.

You have experience building the platforms and tools that empower teams to develop, evaluate, and operate AI agents at production quality. Ideal candidates will have background in one or more of the following areas:

• Data Management & Labeling — annotation platforms, training data pipelines, dataset versioning, data quality assurance, and data governance
• Agent Evaluation & Quality Assurance — benchmarking, automated evaluation, A/B testing, human-in-the-loop review, and feedback-driven model improvement
• Workflow Orchestration & Agent Runtime — multi-step task orchestration, agent observability, debugging tooling, and developer-facing configuration platforms
• Tool / Plugin Ecosystem Management — tool registries, API connector platforms, developer marketplaces, permission management, and security sandboxing

• →Own the end-to-end product lifecycle for AI agent infrastructure: from problem discovery and requirement definition, through development and launch, to post-launch iteration
• →Define product vision, strategy, and roadmap for infrastructure that serves internal AI product teams as primary customers
• →Partner closely with engineering, AI/ML research, and business stakeholders to translate complex technical capabilities into well-scoped, shippable product requirements
• →Establish evaluation frameworks and quality metrics that drive continuous improvement of agent performance, reliability, and safety
• →Build feedback loops connecting production signals (user ratings, task success/failure, escalations) back into model and data improvement pipelines
• →Define platform SLAs, KPIs, and capacity plans; balance scalability and reliability with fast iteration
• →Stay current with the rapidly evolving AI agent landscape and translate emerging capabilities (new models, tool-use paradigms, orchestration patterns) into infrastructure opportunities

• 5+ years of product management experience, with at least 2 years focused on AI/ML products or developer infrastructure
• Demonstrated experience shipping AI products or platforms to production (consumer or enterprise)
• Strong technical acumen: comfortable discussing model architectures, prompt engineering, API design, and system trade-offs with engineers
• Solid understanding of the AI agent development lifecycle: from data collection and annotation, through model training and evaluation, to production deployment and monitoring
• Data-driven mindset with experience defining KPIs, running experiments, and making decisions under uncertainty
• Strong written and verbal communication in English; Mandarin Chinese is a significant plus

• Hands-on experience with AI agent frameworks or platforms (e.g., Claude Skills, Anthropic MCP, LangChain / LangGraph, OpenAI Assistants API, Coze, Dify, or similar)
• Experience building PaaS-level platforms: agent development environments, low-code/full-code toolchains, or multi-agent orchestration systems
• Background in software engineering or technical PM with coding experience
• Experience with responsible AI practices: safety evaluation, bias mitigation, hallucination detection, and guardrail design
• Familiarity with the Web3 or fintech domain
• Understanding of the competitive landscape: familiarity with how leading companies (both in China and globally) are approaching agent infrastructure and evaluation