Skip to content
View fanyang-888's full-sized avatar

Block or report fanyang-888

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse

Pinned Loading

  1. ab-testing-cookie-cats ab-testing-cookie-cats Public

    A/B test analysis of Cookie Cats gate placement (level 30 vs 40). Two-sample proportion tests, Mann-Whitney U, bootstrap CIs, power analysis. Finding: gate_40 caused D7 retention drop of 0.82pp (p=…

    Jupyter Notebook

  2. personal-ai-intelligence-tool personal-ai-intelligence-tool Public

    Sipply — Full-stack AI news aggregator. Next.js + FastAPI + PostgreSQL. 11-stage daily pipeline: ingest → ML score → cluster → LLM summarize → email digest. TF-IDF scorer (Pearson r=0.75), bilingua…

    Python

  3. recommender_system recommender_system Public

    Large-scale news recommendation system on 3M+ user-item interactions (300K users). Item-CF / User-CF collaborative filtering + LightGBM CTR re-ranking. Multi-stage pipeline: embedding retrieval → g…

    Jupyter Notebook

  4. stocktrader stocktrader Public

    Two competing LLM-powered trading agents (Momentum vs Contrarian) + an Evaluator agent. JSON-structured outputs, Yahoo Finance integration. Dual-strategy debate architecture driven by Claude API.

    Python

  5. RFM RFM Public

    End-to-end customer segmentation: RFM feature engineering → K-Means clustering → VIP / frequent / at-risk segments. Top 20% customers generate 74.6% revenue and 85.7% estimated LTV. Includes decisi…

    Jupyter Notebook

  6. UN-Hackathon-GEO-Insight UN-Hackathon-GEO-Insight Public

    Solo 48h · CMU × Databricks × UN OCHA Hackathon 2026. Agentic AI ranks 77 countries by humanitarian funding gap — MCDM Gap Score, ReAct agent with bias guardrails, MLflow 3 tracing, LLM-as-Judge ev…

    Python