YouTube’s New Feature Lets You Code Your Own Watchlist

Finding a YouTube video that actually matches your vibe has been a long-standing challenge for tech-savvy users in Kenya. We’ve all been there: you watch one specific tutorial or a single Gengetone music video, and suddenly your entire feed is hijacked by irrelevant suggestions. YouTube appears to have finally recognized this algorithmic flaw and is rolling out a fix. The video-sharing giant is currently testing an experimental feature dubbed “Your Custom Feed,” designed to give power back to the user and declutter the chaotic homepage experience.

This new test aims to directly tackle the frustration caused by algorithm-driven recommendations that frequently miss the mark. Historically, YouTube’s recommendation engine—driven by complex machine learning—has been reported to misjudge user intentions with annoying frequency. For example, binge-watching a few tech reviews could lead the platform to presume you’re a die-hard hardware engineer, flooding you with niche content you didn’t ask for. “Your Custom Feed” attempts to solve this by moving away from passive signal interpretation to active user input.

For the lucky users participating in this experiment, a distinct “Your Custom Feed” chip will appear right next to the standard “Home” button on the mobile and desktop interface. Unlike the traditional feed which guesses what you like, clicking this new button allows you to input specific, tailored prompts. It essentially functions as a command line for your entertainment; if you want to see a stream of “Kenyan coding tutorials” or “latest Nairobi tech events,” you simply type that into the custom feed. This encourages the system to prioritize exactly what you are looking for in that moment, rather than what the historical data thinks you need.

While it remains to be seen how widely adopted “Your Custom Feed” will be, the shift toward active customization is a potential game-changer for the Kenyan digital landscape. This feature offers a far more efficient alternative to the tedious process of clicking “Not interested” or “Don’t recommend channel” on individual videos—a method that often feels like fighting a losing battle. By allowing users to curate their consumption actively, YouTube is acknowledging that human intent is often more accurate than predictive AI.

This move is part of a broader trend in the tech industry for 2025, where platforms are racing to offer customizable algorithmic experiences. Meta’s Threads was recently spotted testing a similar algorithm-configuration feature, and X (formerly Twitter) is working on deep integration with its AI chatbot, Grok, to allow users to fine-tune their timelines. As these giants compete for screen time, the ability to control the “noise” on your feed might just be the killer feature users have been waiting for.