meet_wani = {
    "role"       : "Final-year AI & Data Science Engineer",
    "college"    : "Suman Ramesh Tulsiani FOE, Pune (2022–2026)",
    "cgpa"       : 9.10,
    "focus"      : ["Growth Analytics", "Credit Risk", "Product Analytics"],
    "experience" : "Data Science Consultant @ Rubixe AI",
    "tools"      : ["Python", "SQL", "Power BI", "AWS EC2", "Streamlit"],
    "ai_native"  : True,
    "open_to"    : "Fintech Product / Growth / Credit roles 🚀"
}

💡 I build end-to-end data solutions — from raw SQL queries on 1M+ transactions to deployed ML apps — that drive real business outcomes.

• 📦 Owned end-to-end analytics pipelines across 3+ capstone projects — from problem framing to stakeholder delivery
• 📊 Built descriptive, diagnostic, and prescriptive analytics solutions; translated complex findings into clear business recommendations
• 🤖 Developed and deployed ML models (classification, regression) on datasets of 50,000+ records with defined KPIs & performance monitoring
• ⚡ Leveraged AI-native tools (Claude, Copilot) to accelerate workflows and improve code quality

PostgreSQL · CTEs · Window Functions · Fraud Detection · Cohort Analysis

• Executed 15+ advanced SQL queries (CTEs, Window Functions, RANK, LAG) on 1M+ financial transactions
• Built fraud risk scoring using CASE WHEN — flagged anomalous transactions 3x above customer average
• Performed 30/60/90-day cohort retention and CLV analysis across 2,000 customers

Python · SMOTE · Random Forest · Power BI · Streamlit · DAX

• Analyzed 7,043 telecom records to identify key churn drivers
• Applied SMOTE; selected Random Forest (86% accuracy) via 5-fold cross-validation
• Built Power BI dashboard with DAX KPIs — Churn Rate: 26.54%
• Deployed Streamlit app with real-time churn probability scores and retention recommendations

NLP · TF-IDF · Cosine Similarity · NLTK · Streamlit · Pickle

• Built a content-based recommendation engine on 20,000+ Spotify songs using TF-IDF (5K features) + Cosine Similarity
• Applied NLP preprocessing (NLTK tokenization, stopword removal, regex cleaning) for lyrical feature extraction
• Deployed interactive Streamlit web app with real-time search, similarity badges, and lyrics preview

⚙️ To enable the snake: Go to your repo → Actions tab → New workflow → paste the YAML below and commit. It auto-runs daily!

# .github/workflows/snake.yml
name: Generate Snake

on:
  schedule:
    - cron: "0 0 * * *"
  workflow_dispatch:

jobs:
  generate:
    runs-on: ubuntu-latest
    steps:
      - uses: Platane/snk/svg-only@v3
        with:
          github_user_name: ${{ github.repository_owner }}
          outputs: |
            dist/github-snake.svg
            dist/github-snake-dark.svg?palette=github-dark
      - uses: crazy-max/ghaction-github-pages@v3
        with:
          target_branch: output
          build_dir: dist
        env:
          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}

🤝 Let's Connect & Build Something Impactful!

"Data is the new oil — I'm here to refine it." ⚗️