Loading...
Loading...
Please wait
You know that “it’s a simple fix” task that eats your entire sprint?
If you liked this episode or if this saved you a sprint: like, subscribe, and share with your team. Comment your worst “simple fix” story! We’ll feature a few next episode!
This episode is about going from “just parse the RSS” to a real system with cron jobs, a database, SSR, caching, pagination, title-matching pain, and a YouTube Data API gotcha where deleted videos still show up and break your counts. We unpack the technical rabbit hole, the product/process mistakes that made it worse, and the practical fixes you can ship today.
SITE https://www.programmingpodcast.com/
Stay in Touch:
📧 Have ideas or questions for the show? Or are you a business that wants to talk business?
Email us at [email protected]!
Danny Thompson
https://x.com/DThompsonDev
https://www.linkedin.com/in/DThompsonDev
www.DThompsonDev.com
Leon Noel
https://x.com/leonnoel
https://www.linkedin.com/in/leonnoel/
https://100devs.org/
📧 Have ideas or questions for the show? Or are you a business that wants to talk business?
Email us at [email protected]!
Highlights
- Why YouTube RSS only returns ~15 items, and when to switch to the Data API
-The sneaky “deleted video” entries that broke episode matching (and the 4-line filter that fixed it)
- Cron + DB to avoid on-request parsing delays, with lazy loading/pagination for perf
- Levenshtein vs. AI scraping for cross-platform title matching (and tradeoffs)
- SSR for SEO: hydration pitfalls, view-source reality checks, and caching strategy
- Process: ticket sizing gone wrong, sprint rituals that would’ve saved weeks, and a fallback plan when APIs fail
- Career bit (Huntober): the highest-ROI job-hunt moves—ask directly for referrals and quantify your wins so AI can actually write a good resume
What You’ll Learn
When RSS is fine—and when you must use YouTube Data API v3
Designing a resilient ingestion path (cron triggers, rate limits, cost control)
Secure API key handling and avoiding accidental exposure
Concrete heuristics for matching episodes across platforms
The “fallback first” mindset when upstream services go down
Stack & Tools Mentioned
Next.js/SSR, Tailwind/CSS (retro radio UI), cron + DB ingest, YouTube Data API v3, Spotify RSS, Levenshtein distance, AI/LLM parsing workflow, lazy loading/pagination, caching.
Chapters
00:00 It’s “simple”… until it isn’t (cold open)
02:00 50 episodes milestone + data-driven intros
03:20 New personal site goals (personas, UX, content routing)
06:04 Rotary-dial content hub idea
07:42 Plan A: “Just use Spotify/YouTube RSS”
08:56 Parsing delays → cron + DB ingest
11:00 Release cadence (Thurs AM CT) & autosync plan
12:07 YouTube RSS ≈ 15 items?!
13:19 Enabling YouTube Data API v3 (the missing step)
14:22 Title matching fails; publish vs. upload date mismatch
16:31 AI scrape workflow vs. deterministic pipelines
17:13 Levenshtein distance for fuzzy matching
18:53 The painful bug: deleted YouTube videos still in API
20:20 Security considerations for API keys
21:08 Retro CSS “radio” UI + Tailwind
23:01 From 2 points to full sprint (scope creep lessons)
24:03 Rate limits, CORS, and API cost control
24:54 SSR for SEO, hydration errors, caching
26:24 Web creativity is back (experimentation talk)
27:29 Sprint Zero / refactor time that saves real sprints
28:24 Resilience: API fallback to RSS
29:18 Perf: lazy loading & pagination
30:01 Tests vs. cowboy deploys (real talk)
31:20 Takeaways: when to keep it simple vs. do it right
36:01 What is Huntober?
37:41 Highest-ROI job hunt move: ask for referrals
39:07 Make AI useful: quantify your work
41:15 Outro