Designing a YouTube Digest for Signal Over Noise
Source: Dev.to
Why I Built This
I follow a lot of YouTube channels around data, analytics, and AI. Over time, it became overwhelming—new videos every day, long talks, podcasts, webinars. Though all were useful, watching them still demanded time and focus.
I tried the usual things, and none of it really solved the problem:
- Saving videos for later
- Creating playlists
- Watching at higher speed (1.5x or 2x)
- Asking an LLM to “summarize this video” one by one
At some point, I realized the issue wasn’t YouTube; it was how I was consuming it. I didn’t need to watch everything—I needed a better way to decide what actually deserves my attention.
What I Wanted Instead
Instead of watching more videos or reading longer summaries, I just wanted to stay informed, quickly scan, and move on.
Goal: Cut the noise & focus on the signal!
- Tell me what’s new
- Highlight ideas, tools, or patterns
- Help me decide whether to watch or skip
If a video introduces something genuinely interesting, I want to know. If it’s repetitive or low‑signal, I’m fine missing it. Automation should handle the volume, and I should only step in when needed.
The Core Idea
The idea was simple: let automation handle the heavy lifting.
Instead of manually opening videos:
- The system picks up newly uploaded videos from channels I follow.
- Passes the video URL directly to an LLM (Gemini).
- The model processes the video and returns a short, structured summary in a format I control.
The focus isn’t on full summaries, but on highlights—what’s new, what stands out, and what’s worth attention. The result is a compact digest I can scan in minutes and decide whether a video deserves deeper time.
How the System Works (At a High Level)
The system runs as a small pipeline made up of independent steps. Each step works on its own, making the system easy to extend, replace, or run independently.
- Identify newly uploaded videos from the YouTube channels I’m interested in.
- Classify each video (shorts vs. long) so the system knows how to handle it.
- For longer videos, pass the video URL to an LLM, which returns a concise summary based on a predefined structure.
- For short videos (less than 3 minutes), capture the title and the link.
- Group everything by channel and compile it into a single digest.
The final output is sent as an email, but it can also be stored for later use or extended further as the project evolves.
Key Design Decisions
- Conscious API usage – YouTube APIs are quota‑based, so the system fetches only what’s necessary and avoids redundant calls.
- Loosely coupled processing steps – Each pipeline component can evolve independently (e.g., video fetching, summary generation, output delivery).
- Signals over noise – Summaries are intentionally short, surfacing only what’s new or meaningful. If something doesn’t add signal, it doesn’t earn space in the digest.
What I Intentionally Didn’t Build
The intent was to solve my problem first and validate the idea before scaling. Therefore, I didn’t:
- Build a UI or dashboard – an email works well for quick consumption; a frontend would add complexity without core value.
- Add a database to park the data.
- Make this a real‑time system – a daily or periodic digest is sufficient.
These decisions kept the system small, focused, and easy to evolve.
What I Learned Along the Way
- Channel subscriptions: I’m subscribed to 74 channels. Seeing that number in the API response explained why things felt overwhelming.
- Gemini cost (Vertex AI): Be cautious with very long videos; a 2 h 40 min video cost $1.
- YouTube Data API quota: Generous free limit of 10 K units per day (each API hit is 1–100 units), which works well for this use case. It’s a soft limit; you can request more with approval.
- Noise in the data: Early summaries were long and not useful. Iterating on the prompt and output structure helped surface the signal.
What’s Next
I’d like to explore the following over the next few weeks:
- Refine how different types of content are handled. Not every video needs the same level of detail, and there’s room to tune this further.
- Track AI cost.
- Look for optimizations.
Final Thoughts
This started as an idea to see if automation could help. As the project evolved, it became a pet project that I now rely on. I don’t watch fewer videos because of this; I just watch the right ones, saving a lot of time and mental energy.
If you’re in a similar situation with YouTube or long‑form content, this approach might be worth trying. If you’ve tried something similar or have ideas to build on this, I’d love to collaborate.
Links
- GitHub repo: (link not provided)