FistBump is a minimalist social media platform designed for children. I built it very early in my programming journey, just as I was learning React and Firebase, so the UI is basic and the design rough. The platform allows children to post content, react to others’ posts, and earn badges or use fun avatars, all in a simplified environment that discourages prolonged use. Safety is central: FistBump uses an AI to detect profanity alongside a growing dictionary of known bad words, which expands as users report inappropriate content. Users receive strikes for violating rules, and accumulating three strikes results in a temporary ban from posting. The platform deliberately avoids private messaging and algorithmic feeds, keeping interactions transparent and public while minimizing addictive features.

Despite these measures, the moderation system has limitations. While the AI and word dictionary filter common inappropriate language, they do not reliably detect suggestive or negative content. Users can report posts or comments that bypass automatic detection, but more advanced threats—such as grooming or predatory behavior—would require additional safeguards. Real-world deployment would also need age verification with parental consent to prevent adults from posing as children. FistBump reflects my thoughtful approach to child safety and lightweight social interaction, but it also highlights the complexities of building secure digital spaces for younger users and the ongoing need for robust, layered moderation.