Skip Clips

Hey everyone,Quick progress update on Skip Clips after a busy couple of months.

For those who don’t know, Skip Clips is an AI-powered platform that summarises long-form podcasts and videos so you can find the useful parts without listening to everything. We’ve been steadily building out the archive and features

Here's what shipped recently:

1. Redesigned homepage
Moved away from long rows of shows that made sense when we had only a few shows. The homepage now focuses on recent/featured episodes and has a clearer search bar. Also cleaned up a bunch of smaller UI issues.

It's still far from perfect but a step in right direction.

2. Hover descriptions
You can now hover over episode thumbnails to see a short summary without clicking through. Something small but meaningful.

3. Show-level performance analytics, potential new opportunity.
Added analytics on individual show pages (powered by YouTube data for now). Metrics include core audience size, 30-day views gain, hit rate, engagement rate, comment/like ratio, and growth graphs.

This is also an avenue that we are considering exploring to offer more advanced analytics on topics, keywords, shows and guests.

4. Launched the blog
Started publishing longer updates and articles here:
https://skipclips.com/blog/

Not much here at the moment, but a start is a start!

5. Growing Pains: Stronger processing pipeline

Many hours the last month were spent rebuilding the processing pipeline between the new episodes being fetched via the YouTube API, downloaded, uploaded onto the transcription server, transcribed, synced back up and finally a summary generated.

In this past for many months this was just witchcraft of many independent scripts running together with no visibility on status or troubleshooting issues, which there is many. At scale doing something right 99% of the time still leads to issues that requires effective logging and tools to track

Other similar changes were made throughout the back-end, particular to do with scaling. Shortcuts that worked for smaller numbers of items (sub 1000), now no longer work with tens of times that amount and delaying fixing it was only causing more pain!

6. Category tagging (still in progress)
This is currently one of the bigger challenges. We have over 50,000 unique tags, many of them very similar.

We’re working on organising them into proper categories so people can browse by topic instead of only by show. It’s still quite manual at the moment, but once it’s done it should also support future features like following specific topics. AI could be used, but at the moment it's more developing the category tree first and getting a better idea of the tags.

Don't blindly use automation to solve problems you don't full understand!

7. New shows added
Expanded the catalogue with several more shows, including American Thought Leaders, Into the Impossible with Brian Keating, Making Sense with Sam Harris, On with Kara Swisher, The Dr. Phil Podcast, and others.

Currently at 91 shows as of posting, 100 would make a nice number. Quality and suitability is however the aim over shear numbers.

What’s next

• Get category tagging into production

• Expand the number guest profiles, hopefully start experimenting with AI to see what it can do. Reliability is still very important.

• Publish more blog posts. Trying to extract and write about unique useful data obtained from the platform would be nice.

• Start laying the groundwork for user profiles, notifications

Traffic is still very low while we work through indexing and structure issues, but the archive and feature set keep growing. The data side feels increasingly solid.

Would love any feedback, especially around the tagging problem or what features would actually make this more useful for people.

Full Update:
https://skipclips.com/blog/august-2026-development-update

Thanks for reading.
Squawki

Thank You all for showing interest in this project.

The last couple of months have been a blur with push to complete milestones and push Skip Clips closer to being a fully functional platform.

Some of the completed objectives since the last update include:

• Adding of over 19 new shows and over 8200 hours of new content

• Fully switching over to the YouTube API to import new episodes and gathering of statistics to replace the now defunct Spotify API. This is tied in with the new 'Performance Insights' for episodes.

  

• Creation of Guest & Artist Profiles functionality which are interlinked to shows and episodes

  

• Improving the search bar functionality (Live) and development of an 'Advanced Search' (not officially released) functionality to search through descriptions, topics, meta-data and transcripts!

• New 'Trending' page with the latest trending episodes based on YouTube statistics.

• New section to suggest other related episodes based on topic tags.

• Further tweaking of the Transcribing pipeline to get new episodes published faster without a human in the loop.

• Continued polishing and fixing of UX features

With some major features out of the way, the shift for the next couple of months is focused around further automating functions including creating Guest Profiles, Guest Detection and 'Related Media' such as books.

Additionally we are looking at different strategies to organically promote the platform, such as contributing on Reddit and other forums. A task I found easier said than done!

As always, thank you for your continued interest and showing support. It is genuinely appreciated. If you have any ideas or feedback, please reach out.

All the best,

Squawki
Founder

Thank You for taking the time to read this post and exploring the Skip Clips platform.

Skip Clips is an online platform designed with the primary goal of unlocking knowledge hidden within often hours long podcasts and videos. The idea was born after forever not knowing what I was going to expect listening to a podcast or if I was just going to waste my time.

If only their was a tool to get a summary of the entire podcast and show. Far easier said than done! A task that grows in complexity every time I think about it.

After over a year the platform has taken shape with over 37,000 hours of content transcribed and summaries created across 53 shows. More added every day!

This February will be a another busy month building out the interface, the current plan is:

• Fully phasing out the outgoing Spotify API and switching over to the YouTube API.

• Improving the statistics, analytics and tagging with 'Trending, Popular, Viral' badges of episodes. Later in the month investigating dedicated pages (ie This Weeks Viral Podcasts)

• Mass import of remaining episodes for shows with the new transcribing pipeline leveraging Vast.AI

• Tweaking up some minor UX issues (ie black on grey text) in the interface

• Scoping the feasibility of an 'advanced search'

• Further work finishing up the Artist pages and respective profiles. Later expand this feature onto guest profile.

• Promoting the platform on Social Media (X, Reddit)

If you have any feedback, it is always welcomed and appreciated.

Kind Regards,

Squawki
Founder