Available for hire · AI Product Manager

Hi, I'm Betty Xu

AI Product Manager · Agent Products · Model Evaluation · AI Coding

Master's Student in Information Resources Management at Nanjing University

Former Product Intern, LLM Applications · TikTok, ByteDance (Jan 2026 – Sep 2026)

AI Product Management Agent Product Design Model Evaluation Systems Bad Case Analysis Prompt Engineering AI Coding Delivery

View Portfolio View CV

Agent AI Product

Smart Choice Agent | Structured, Explainable Decisions

Turning vague intent into ranked options, constraint analysis, evidence records, and auditable traces

Smart Choice Agent turns one recommendation into an inspectable decision process: it understands intent and hard constraints, generates candidates, ranks them, explains the trade-offs, and records the trace for replay and expansion.

My role: Independently designed the decision product framework, multi-agent orchestration, deterministic ranking engine, diet-domain plugin, local web UI, and auditable trace model.
Result Delivered: The project is public and locally runnable. Diet recommendation is the first complete domain, while the generic decision workbench and additional domains remain in progress.

Visual material pending

AI Agent AI Product Design

Insight AI

Helping people judge health information—not just receive answers

Insight AI does more than judge one health claim. It connects evidence explanation, short practice, and progress review so users can build a judgment method they can reuse.

My role: Individual graduation project; independently owned product positioning, user needs, core features, Agent workflow, prototype design, and validation planning.
Result Validating: Completed the product concept, core feature flows, Agent prototype workflow, and validation framework; user outcomes remain under validation.

Visual material pending

Core Capabilities

Connecting product thinking, model evaluation, and hands-on delivery

AI Agent Product Design

Built an Understanding → Retrieval → Reasoning → Reflection workflow with profile-based intervention strategies

See evidence

Model Evaluation & Experiment Design

Designed a three-group controlled study around susceptibility profiles and four evaluation dimensions

See evidence

Specification-Driven AI Coding

Used specifications to define behavior and failure boundaries across implementation, integration, and launch

See evidence

Project Runnable Eval Demo · Real-model integration in progress

Agent Eval Studio | Release-Oriented Agent Evaluation

Built an evidence-first Agent evaluation workbench with 60 samples across 10 task types, a five-dimensional rubric, A/B comparison, bad-case analysis, and quality, cost, P95 latency, and safety gates.

Agent EvalAI ProductEvaluation

Tool Open Source

Repo Storyteller

Turns repositories into evidence-first project explanations, separating facts, inferences, and unknowns while detecting forks, tutorials, and implementation boundaries.

Codex SkillCode AnalysisEvidence

Tool Open Source

Job Search Skills

End-to-end job search skills: career positioning, JD analysis, resume tailoring, interview prep, and offer decisions — with an AI product orchestrator.

Job SearchSkillsAI Product

Notes

Product thinking, hands-on AI practice, product analysis, and reusable tools and skills.

When AI Becomes a Product Capability: Start with the Scenario, Not the Model

A developing note on evaluating AI product opportunities by starting with user tasks, problem frequency, and success criteria before choosing where model capabilities belong.

2026-08-09

Product Thinking AI Product Opportunity Assessment

Hi, I'm Betty Xu

Smart Choice Agent | Structured, Explainable Decisions

Insight AI

Core Capabilities

AI Agent Product Design

Model Evaluation & Experiment Design

Specification-Driven AI Coding

Open-source tools

Agent Eval Studio | Release-Oriented Agent Evaluation

Repo Storyteller

Job Search Skills

Notes

When AI Becomes a Product Capability: Start with the Scenario, Not the Model