AI Workflow Engineer (early career) Paris, France
• Python • Streamlit • Workflow Automation • RAG

I started my AI career journey in October 2025. I build small, practical tools in Python with clean Streamlit UIs — designed to demo in under a minute (sample inputs included).

Currently studying Harvard edX: CS50’s Introduction to Computer Science + CS50’s Introduction to Artificial Intelligence with Python.

I’m documenting builds and walkthroughs on YouTube: @PonyRivers-x7i

• Channel: https://www.youtube.com/@PonyRivers-x7i
• Demo: Portfolio Assets Finder → https://youtu.be/MYZJR9mZTZ0
• Demo: NFL Betting Engine Case Study → https://www.youtube.com/watch?v=xI88nChjKWU

If you watch a demo, feedback is welcome — I’m improving clarity and demo flow with each release.

Structured search → review results → export timestamped run folders (CSV/JSON), optional downloads, plus an Instagram caption pack (links-only by default).
Live: https://mrponyrivers-portfolio-asset-finder.streamlit.app/
Repo: https://github.com/mrponyrivers/portfolio_asset_finder
Demo: https://youtu.be/MYZJR9mZTZ0

Categorizes bank statements into business/personal/tax buckets with clean exports for reporting and accountants.
Live: https://mrponyrivers-tax-buckets-app.streamlit.app/
Repo: https://github.com/mrponyrivers/tax-buckets-app

Practice betting fundamentals: odds (decimal/American) → implied probability → edge/EV → Kelly sizing, plus Trainer drills + CSV backtesting with bankroll curve + drawdown.
Live: https://mrponyrivers-bet-decision-trainer.streamlit.app/
Repo: https://github.com/mrponyrivers/bet-decision-trainer
Demo: https://youtu.be/p4s3mTzzzJU

📉 Line Shopping + CLV Tracker (Streamlit) Compare sportsbook odds, select best prices, and track CLV over time.

Live App: https://mrponyrivers-line-shopping-clv-tracker.streamlit.app/ Repo: https://github.com/mrponyrivers/line-shopping-clv-tracker Demo: https://youtu.be/ghMJ3IXv45Y---

• 📘 Powerbook Training App
  Interactive Streamlit app for betting math drills

A training app built with Python and Streamlit to help users practice the core math behind betting markets and decision systems.

Skills trained:

• Implied Probability
• Expected Value (EV)
• No-Vig Probability

Features:

• interactive drills
• instant answer checking
• step-by-step explanations
• score tracking
• beginner-friendly practice flow

Links:
🔗 Live App: https://mrponyrivers-powerbook-training-app.streamlit.app/
🔗 Repo: https://github.com/mrponyrivers/powerbook-training-app
🎥 Demo: PASTE-YOUR-VIDEO-LINK-HERE

• 🗓️ Itinerary Calendar Builder — paste itinerary text → review/edit → export WORK/TRAVEL/HOLD .ics calendars (RunID batch delete)
  Live: https://mrponyrivers-itinerary-calendar-builder.streamlit.app/
  Repo: https://github.com/mrponyrivers/itinerary-calendar-builder

• 📄 RAG PDF Assistant — local PDF Q&A with source transparency + page citations
  Live: https://mrponyrivers-rag-pdf-assistant.streamlit.app/
  Repo: https://github.com/mrponyrivers/rag-pdf-assistant

• 📄 Resume Extractor — extract structured fields from PDF/TXT → JSON + Markdown report
  Live: https://mrponyrivers-resume-extractor.streamlit.app/
  Repo: https://github.com/mrponyrivers/resume-extractor

• 📈 Odds EV Calculator — implied probability, fair odds, EV, and Kelly stake sizing
  Live: https://odds-ev-calculator-ponyrivers.streamlit.app/
  Repo: https://github.com/mrponyrivers/odds-ev-calculator

• 🏈 Sports odds + decisioning workflow (NFL POWERBOOK) — end-to-end pipeline from odds normalization → implied probability → EV/Kelly decisioning → logging/settlement → CLV tracking
  Repo: https://github.com/mrponyrivers/nfl-betting-engine-case-study
  Demo video: https://www.youtube.com/watch?v=xI88nChjKWU

• Local-first workflows (no external LLM required for core functionality)
• Clear outputs (downloads: ICS/CSV/JSON/Markdown/XLSX)
• Recruiter-friendly demos (works in ~30 seconds with sample input/files)
• Turning messy real-world inputs into reliable outputs (structured data, exports, repeatable workflows)
• Shipping “portfolio-ready” apps: docs, demo flow, and deployment

Python • Streamlit • Git/GitHub • Workflow Automation • ETL • API Integration • Data Cleaning
CSV/JSON Exports • File-system Pipelines • RAG (Document Q&A) • Deployment (Streamlit Cloud)

Python • Streamlit • pandas • pdfplumber • openpyxl • PyPDF • sentence-transformers • embeddings retrieval

GitHub: https://github.com/mrponyrivers