Python Developer for Snapchat Archiver Fix

Need Experienced Playwright/Python Developer to Fix Snapchat Web Chat Archiver (Text Extraction + Virtualized DOM)
Project Overview
I’ve built a personal Snapchat Web backup/archive tool using Python + Playwright. The tool already handles login persistence, scrolling, media extraction, media downloading, checkpoints/resume, transcript generation, and anti-bot delays.
The idea of the script is to open chromium, select a chat and download all the media and texts within the chat and save it as a html file which can be viewed in the chat format.
Current status:
• Media extraction/download works correctly
• Browser automation works correctly
• Login/session persistence works correctly
• Infinite upward scrolling works partially
• HTML transcript generation works
The remaining issue is TEXT MESSAGE extraction from private chats.
What the tool currently does:
• Opens Snapchat Web in a persistent browser session
• Navigates into a private chat
• Scrolls upward gradually to load older messages
• Extracts messages from DOM
• Downloads media before CDN expiry
• Builds local HTML transcript
• Maintains checkpoint system to avoid duplicates
Main Problem:
Many text messages are not detected properly and are not saved. MEDIA extraction already works. The task is ONLY about reliable TEXT extraction and stable scrolling/deduplication. I want you to refractor the code so that the tool can effectively extract texts in the chats as well and save it as a html file.
Technical Stack
• Python
• Playwright
• Asyncio
• Local HTML transcript generation
I need an experienced automation/scraping engineer who can:
1. Analyze Snapchat Web private chat DOM behavior
2. Build a reliable text extraction system
3. Handle React virtualized lists properly
4. Prevent duplicate collapsing
5. Ensure all visible text messages are archived
6. Stabilize upward scrolling/history loading
7. Improve message identification logic
8. Make extraction resilient to Snapchat DOM updates
Requirements
Deliverables
I expect:
• Fully working Python code main.py that can generate a transcript.html file which has texts from both parties with timestamps and media.
• Clean integration into existing script
• Stable extraction of text messages
• Proper deduplication
• Reliable scrolling/history loading
• Comments/documentation explaining fixes
What I Will Provide
• Current Python source file
• Logs
• Sample transcript.html
• Current extraction logic

Budget:
I've already completed the project, only the text part is remaining. So the Quote has to be less than 2500.

Check files before Quoting.

Deadline: Under 40 hours.