Skip to content
View anthonydenis01's full-sized avatar
🎯
Focusing
🎯
Focusing
3 followers · 1 following

Achievements

Achievement: YOLO

Achievements

Achievement: YOLO

Highlights

  • Pro

Block or report anthonydenis01

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
anthonydenis01/README.md

Anthony Denis

I build AI agents that automate import logistics operations.

I work inside the Import VIP desk of one of the world's largest container carriers. That's where you learn where freight operations bleed time: invoice matching, container exception hunting, carrier chasing. I build the agents that do that work — grounded in how ops actually run, not how software vendors imagine they run.

🔴 Live demos

Project What it does Try it
Middle Watch 🆕⭐ On-water exception monitoring — routing changes, vessel swaps, dwell, ETA drift, transit vs routing guide. Agentic tool-calling + MCP, field-level evidence on every line. 40k containers in 2.3s, 100% recall / 0% false positives Demo · Code
Import Ops Agent Full daily verification — carrier sites, terminal holds, invoice audit, prioritized exceptions Interactive demo
Freight Document Intelligence RAG Q&A over full shipment doc packs — page-level citations, cross-doc reconciliation, compliance audit (44/44 tests) Demo · Code
Import Desk Automation Invoice↔booking matching (97%), overcharge detection, AR audit Dashboard · Code
Container Visibility Vessel delay detection, 48h detention-risk alerts Dashboard · Code
Carrier Scorecard Carrier reliability rankings by lane, switching-savings estimates Dashboard · Code

All demos run on synthetic data. Deterministic pipelines with a Claude reasoning layer on top — the math never depends on the LLM.

How I build

Python · Claude API (tool use) · MCP · RAG · Pandas · GitHub Pages — agents designed for adoption: they output what ops teams already use, flag rather than auto-act, and measure hours saved, not model metrics.

Background

🚢 Import VIP desk @ one of the world's largest container carriers · 📍 Miami · 🇫🇷 French

📫 LinkedIn

Pinned Loading

  1. carrier-scorecard-agent carrier-scorecard-agent Public

    AI agent that ranks carrier reliability by trade lane, flags underperformers, and quantifies the savings of reallocating volume.

    Python 1

  2. container-visibility-agent container-visibility-agent Public

    AI agent that tracks container and vessel delays, flags detention risk 48 hours in advance, and drafts mitigation actions for logistics operations.

    Python 1

  3. import-desk-automation-agent import-desk-automation-agent Public

    AI agent that automates invoice-to-booking matching, overcharge detection, and AR auditing for import logistics operations.

    Python 1

  4. import-ops-agent-demo import-ops-agent-demo Public

    Interactive demo: AI agent that verifies containers across carrier sites, terminal systems, and invoices for import operations.

    HTML 1

  5. freight-document-intelligence freight-document-intelligence Public

    Ask your shipping documents anything - cited Q&A, cross-document reconciliation and compliance checks for freight document packs. API-first, human-in-the-loop, synthetic data.

    Python