I let AI handle my Plex transcode and saved 8.5TB without lifting a finger

Published Sep 21, 2026, 7:30 AM EDT Patrick Campanale has been in the tech space for well over a decade, specializing in PC/gaming news and reviews, as well as maker-focused products to build small businesses. With a start in technology back in 2010 surrounding the Palm/webOS ecosystem, Patrick spent his formative years developing mobile applications as well as blogging for various publications, eventually leading to starting his own website in 2014. After running a technology blog for a few years, he stepped out of that role and into the world of high-end custom PC manufacturing and building, with a focus on YouTube video production and overclocking. Then, in 2017, Patrick joined the 9to5Toys team as an editor/writer/reviewer with over 14,000 articles being published during his time there, ranging from deals and roundups to in-depth reviews on the latest technology, video games, 3D printers, and more. In his free time, Patrick loves to create projects from wood using various robots and methods, including leveraging the technologies of CNCs and lasers. If Patrick isn't working on a computer or playing video games, he's likely in his 2-car garage workshop creating something unique. In addition to all this, Patrick is also the media director at his local church where he feels God has called him to serve, and he loves every minute of it. My Plex library had grown far too large, and it was time to downsize it. I could delete stuff, sure, but I wanted to keep most of what was there. So, I turned to converting my entire library to H.265, and it saved me nearly 10TB of space. Downsizing my NAS meant downsizing my library Deleting old movies and TV shows only got me so far I recently made the move from running my old enterprise-grade NAS with 12 drive bays and 80TB of storage to a UGREEN NAS with 6 drive bays populated with three drives and a total of 40TB of storage. This downsize saves me quite literally hundreds of dollars per month in electricity, but it didn't come without its downsides. My Plex library is quite sizable, measuring at well over 40TB before the move. So, I started off by deleting things that nobody had watched in years. I've been running my Plex library for about 6 years at this point, and some content had been on there since the beginning and never watched. I use Codex for everything these days, and I had it integrate with Tautulli to find the media nobody had been watching and give me a list of delete candidates. A similar thing could be accomplished with Maintainerr, I just chose to do it with Codex. From there, I nuked tens of terabytes of media from my server that I just really didn't "need." I never did it before because I had 80TB of usable space, so I just kept everything. Once I deleted everything I didn't need, I was down to about 30TB. That might sound like a lot, considering I had 40TB to work with, but it still wasn't comfortable enough for me. So, I decided to take the next step: transcoding my entire Plex library. The last time I tried transcoding it took forever and was a ton of work I had Gemini writing my ffmpeg scripts and it wasn't pretty Credit: Patrick Campanale / How-To Geek In early 2026, I spent some time transcoding some of my movies from H.264 to H.265. It was a pain in the neck and a complete mess, to be honest. I definitely got the job done, but it was complicated. I was using AI to help me come up with the ffmpeg scripts, but it was all manual. I would feed my needs and setup to Gemini (the AI I was using at the time) and it would give me a script to use. I'd test it until it failed, then try to debug it with Gemini. There was a lot of back-and-forth here, and that was the most painful part of all. Things would look like they were working fine, and then fail overnight and waste hours upon hours of time that could have been spent transcoding. Overall, it worked but it didn't work how I wanted. I was working with a massive library and only regained 7TB of space. This time around though, when I started with right around 30TB of media, I went a different route—I had Codex handle it from start to finish. Shrinking my library this time was simple and easy H.264 to H.265 was easy when I let Codex do the heavy lifting One of my favorite features of Codex is that it can run scheduled tasks. So, when I set out to transcode my library this time, I decided to hand the task off to Codex. Codex used SSH to configure all of my nodes that I would use for Tdarr, as well as building custom ffmpeg scripts and plugin flows inside of Tdarr for each node. In total, I ran five nodes to get through the transcoding as fast as possible. I chose to transcode to H.265 because my Plex server now runs on a newer platform that's capable of handling HEVC content natively. Plus, H.265 takes up considerably less space than H.264. Some of the nodes used NVIDIA CUDA, while others used integrated Intel graphics. This was one of the biggest problems when I set up Tdarr the first time, as the plugin flow was a pain. Codex handled it like a champ though. I basically told Codex what I wanted and it configured everything for me. From there, I told Codex to monitor the transcode status every 15 minutes. So, every 15 minutes, it would update the thread with the current progress and also flag any errors. Here's where Codex started to really pull its weight—I had it fix the errors automatically for me. I wasn't worried about losing media at this point, I had a backup on my old NAS and I could re-rip anything I needed to. So, during the 15-minute checks, if Codex saw an issue, it would fix the script and re-queue whatever failed. I basically let Codex run the transcode for about 72 hours, as that's how long it took for everything to process through. At the end of it all, I regained 8.5TB of space just from Codex transcoding my media. Overall, I shrank my media library from around 40TB to around 16TB, which left me with plenty of headroom for the server to grow over time again. AI has proven its usefulness in my homelab time and time again While AI is far from perfect, it's definitely great at a good many things. Having it run this transcode was like having an IT pro sitting in my office monitoring it every 15 minutes without taking a break for sleep or food. I definitely couldn't have accomplished this without Codex's help, at least, not in the time I did it in and with as little frustration as I had.