Python Developer Needed: Weekly Transcript Scraper to Word Doc (YouTube, Podcasts, Web)
Freelance Developer Needed: Weekly Transcript and Article Collector (One Document Output, One-Off Build)
Overview
I want a simple automation that tracks a list of specific people (I will provide the names) and, once per week, searches for any new content featuring them (YouTube, podcasts, blogs/articles and similar). The tool should pull the full transcript or article text where available and compile everything into one single document per week with clear headings.
I am a novice at coding. I use VS Code at a basic level. I do not use GitHub and I do not want a complex setup. I also do not want ongoing help or retainers. I want a one-off build with a clean handover so I can run it myself.
What you will build
Core output: Every week, generate one Word document (.docx) saved automatically to a folder on my drive (I will choose the folder path).
Required behaviour:
•I will provide a list of approximately 35 prominent AI voices (e.g. Andrej Karpathy). The list may grow slightly over time, so the solution should make adding names easy.
• The script runs weekly (manual run is fine, or optional scheduled run).
• It searches for new items featuring those people across:
o YouTube (interviews, podcasts, talks)
o Podcasts (prefer RSS-based discovery where possible)
o Web articles/blogs (interviews, posts, transcripts)
• For each item found, it pulls:
o Full transcript
o Podcast transcript text
o Full article text (for web pages)
• It compiles everything into one Word document each week with clear structure, for example:
o Person name
Source type (YouTube/Podcast/Article)
Title
Date published
Link
Transcript/article text
• It must avoid duplicates:
o If something was already included in a prior week, do not include it again.
• It must have basic logs:
o What it found, what it downloaded, what failed (and why).
My setup and constraints
• Windows PC
• I want each weekly file saved to a folder on my drive
• I do not want to manage databases manually or deal with complex developer tooling
• I want a simple “run” process (ideally one command or a double-click file)
Deliverables
• A working solution on Windows that I can run weekly
• Generates one Word document per week in the chosen folder with the format above
• De-duplication working across weeks
• A simple configuration file where I paste and update the list of names
• Beginner-friendly handover instructions:
o How to install/run it
o How to add/remove names
o What to do if a source fails
• Optional (nice-to-have): a shortcut or Windows Task Scheduler setup so it runs automatically weekly.
Important requirements
• One-off build, not ongoing support
• Clear handover instructions written for a beginner
To apply
Please send:
1. A short description of similar automation / transcript ingestion work you’ve done
2. Your recommended approach (high level)
3. Your fixed price
4. How quickly you can deliver a working prototype
5. Any risks you see (rate limits, missing transcripts, etc.)
Note on tooling
VS Code is fine. Please deliver it so I can run it without needing GitHub or anything complex (for example, a simple run script plus clear setup instructions).
Apply tot his job
Apply To this Job