AV WTP Tracker Web Application

I have a calibrated research model that converts textual news and policy statements about automated vehicles (AVs) into dynamic estimates of the public’s willingness-to-pay (WTP). What I now need is a robust, public-facing web application that takes that model out of my notebooks and puts it in front of everyday users.

Scope
• Deploy the full application on AWS – that is my preferred and only hosting environment.
• The ingestion layer will receive regular CSV drops containing the raw AV text data. Your code should validate each file, pass it through the existing model (Python notebooks and a pickled scorer are ready), and write the resulting WTP estimates with timestamps to a database you set up.
• On the front end, visitors must be able to:
– Filter the data by date range, topic tags and geography.
– Generate custom PDF/CSV reports on demand.
– Explore WTP movements through responsive, interactive charts (think Plotly or D3).

Deliverables
1. Clean, well-documented source code (backend, frontend, and infrastructure as-code for AWS).
2. Deployed live site with secure HTTPS, autoscaling, and nightly backups.
3. One-click admin interface for uploading new CSVs and monitoring model runs.
4. README covering setup, deployment and maintenance steps.
5. Short video walk-through confirming all three user interactions work as specified.

Acceptance criteria
• A new CSV upload appears in the dashboard within five minutes and its WTP results are visible in the charts.
• All filters, report downloads and chart interactions work on Chrome, Firefox and Edge, desktop and mobile.
• Lighthouse performance score ≥ 80 on the main page.
• No critical or high vulnerabilities flagged by OWASP ZAP scan.

If this sounds like your wheelhouse, let’s discuss timelines and any clarifications you need.