I Tried Editing Videos on Linux. Here's Why Kdenlive Fi…

My experience with video editing on Linux, losing an hour of work, almost building my own editor, and finally discovering Kdenlive.

I've been creating content for the last two months. Not a lot, but I think I've produced at least ten videos so far. Some are shorts, while others are long-form videos in the 20–25 minute range. I also have a bunch of recorded content waiting to be edited and published, and if there's one thing I've realised after getting into content creation, it's that recording is only half the battle.

Initially, the biggest challenge is getting comfortable in front of a camera. You have to get over that awkward feeling of talking to a lens, hearing your own voice, and occasionally forgetting what you were supposed to say. But once you get past that, there's another frustration waiting for you, and that's video editing.

I wouldn't call editing the hardest skill to learn, but it definitely takes time, not just to learn how to use an editor, but also to develop a taste for what makes a video good. I don't think I have that taste yet, and although I've watched some really well-produced videos and would love to implement similar things in my content eventually, that's not my focus right now. First, I want to get comfortable creating content consistently, learning basic editing, and getting videos out there. The fancy stuff can come later.

All I wanted was a basic video editor

The thing is, my editing requirements aren't even that complicated. For most of my videos, I just need to cut out sections where I mispronounce something, make a mistake, or go completely off-script, add captions, remove background noise (especially the ceiling fan that somehow manages to become the main character in my recordings), occasionally add some basic transitions or effects, and finally export the video and upload it.

That's it. I'm not trying to produce a Christopher Nolan film. I'm just a software engineer trying to make some videos about engineering and the things I'm building, and yet, even this basic workflow turned into a nightmare.

My unfortunate introduction to OpenShot

I'm on Linux, and finding a good, free, open-source video editor isn't as straightforward as I initially expected. There are a few popular options, including OpenShot and Kdenlive, but unfortunately, last month I decided to go with OpenShot, and after using it for a while, I've come to a very simple conclusion.

OpenShot is shit.

At least, my experience with it has been. In fact, right before writing this blog, I was busy uninstalling it from my system because that's how frustrated I was with the entire experience.

Let me explain what happened today. I was editing my second long-form YouTube video, where the raw recording was around 35 minutes, and I wanted to bring it down to roughly 20–25 minutes. It took me around 45 minutes to an hour to make all the cuts, remove unnecessary sections, and get the video into a somewhat publishable shape. Everything was going fine until I reached the audio processing part, and that's where the actual nightmare started.

A ceiling fan, Python, FFmpeg, and 20 crashes

One problem with my recording setup is background noise. My ceiling fan gets picked up by the microphone, and I wanted to remove that noise before publishing the video. Unfortunately, OpenShot doesn't provide the kind of native background noise reduction I was looking for, so being a developer, I did what developers usually do when a tool doesn't solve their problem.

I built another tool.

It's called Hush, a small tool I built using Python and FFmpeg to process audio and reduce background noise using AI. The workflow I had in mind was pretty simple: finish editing the video, extract the audio, process it through Hush, import the cleaned audio back into the editor, and finally export the video. Simple enough, right? Well, apparently not.

While trying to separate the audio from my video, OpenShot started crashing repeatedly, and I counted at least 20 crashes during that editing session. I'm not even exaggerating. Every time I tried to get something done, there was another crash waiting for me, and then came the final one, which somehow resulted in me losing all my edits. I couldn't recover the work I had spent almost an hour doing.

Imagine sitting there, making dozens of cuts to a long video, getting everything ready, and then watching the editor throw all your work away. I was genuinely pissed, and at that point, I had almost given up on the idea of using OpenShot for anything ever again.

At one point, I thought of building my own editor

This is where things got interesting because I was so frustrated with the available editors that I seriously started considering building my own video editor. Not from scratch, obviously, but my plan was to find a bare-bones, open-source video editor, fork its repository, and use Codex to implement the handful of features I actually needed.

I even started looking for editors whose source code I could download and modify. I didn't need a massive editing suite with hundreds of features. I just wanted something that could cut clips, manage audio tracks, reduce noise, and export videos without crashing every five minutes, and for a moment, this sounded like a perfectly reasonable engineering solution.

Then I realised what I was doing. I had started my day wanting to edit a YouTube video, and somehow I was now planning to fork an entire video editor.

Classic developer behaviour, lol.

Even with coding agents, getting a usable editor running with my required features would probably take another three to five hours, if not more, and I still had a video to publish. So instead of jumping into another unnecessary side project, I decided to give the existing options one last chance and explore another editor.

And then I discovered Kdenlive

Enter Kdenlive.

I had come across it earlier but never really tried it. After the OpenShot disaster, I downloaded it and started exploring, and holy shit, why didn't I try this earlier?

Kdenlive is genuinely one of the best pieces of software I've used for video editing on Linux. The interface felt much better, the timeline editing was smooth, and it had pretty much everything I needed without making me fight with the software to get basic things done. Most importantly, it had proper audio effects, including noise reduction options, which meant I could finally process the background noise directly inside the editor without having to export my audio, run it through another tool, and import it back.

Although I had already built Hush, having the option to do everything inside one editor made the workflow so much easier. Kdenlive also has a bunch of useful plugins, effects, and features that I haven't even explored properly yet, but even with the little time I spent using it, I could already see how much better the experience was compared to what I had been dealing with for the last month.

And the best part? It took me around 15–20 minutes to get comfortable with the basic workflow.

Obviously, mastering an editor takes time, and you might spend hours or weeks learning the advanced features, effects, transitions, colour grading, and everything else. But for someone who just wants to make basic cuts, work with audio, and export videos, the learning curve felt pretty manageable, and I didn't have to spend an entire day watching tutorials just to understand how to use the timeline.

I edited the entire video again in roughly 45 minutes, and this time there were no crashes, no lost edits, and no fighting with the interface. I could just focus on editing, which is literally all I wanted from a video editor in the first place.

The tool shouldn't become another project

One thing I've noticed as a developer is how easy it is to fall into the trap of building a solution for every inconvenience. Something doesn't work? Let's build it. An existing tool is missing a feature? Let's fork it. A workflow feels slightly annoying? Time to create another side project.

And sometimes, that's great because a lot of interesting projects start exactly this way, but sometimes you just need to finish the original task instead of creating five additional tasks around it.

I almost spent an entire evening building a video editor because I was frustrated with another video editor, all while having a perfectly good alternative sitting there, waiting to be installed. I'm glad I explored Kdenlive before going down that rabbit hole because otherwise, I would probably still be debugging some random timeline rendering issue instead of writing this blog.

If you're a developer creating content on Linux, just try Kdenlive

This is pretty much the reason I wanted to write this blog. If you're a software engineer planning to start creating content, especially on Linux, I would strongly recommend giving Kdenlive a try because it solves pretty much all the basic editing requirements without forcing you to spend hours figuring out the software itself.

On macOS and Windows, you have plenty of established editing options, while Linux has good options too, but finding one that fits your workflow can take a little experimentation. For me, Kdenlive solved pretty much everything I needed, and I genuinely wish I had discovered it earlier instead of wasting so much time dealing with OpenShot.

And if you're just starting out, don't worry too much about producing perfectly edited videos. Learn how to record, make basic cuts, clean up your audio, add captions, and publish your content. You can develop your editing taste along the way, and I'm still figuring that part out myself.

For now, I'm just happy that I can finally edit a 25-minute video without accidentally turning it into a five-hour software engineering project.

And yes, OpenShot is officially uninstalled.

Just pick Kdenlive and start creating.