Built, deployed, and shipped 5 image tools solo. Now the hard part: getting users.
Fair question. remove.bg and TinyPNG are excellent single-purpose tools — I still use both. My pitch is different: One suite instead of 5 tabs Bulk mode built-in (20 files at a time on most tools) No signup, no watermark, no daily limits Clear no-storage policy — files are processed in memory and discarded Not trying to beat the incumbents at their specialty — just trying to be the tool that…
Fair question. remove.bg and TinyPNG are outstanding single-purpose tools that the author still uses. The pitch here is different: a comprehensive suite of image tools instead of multiple separate applications. The suite features bulk mode (processing up to 20 files at once), no required signups, no watermarks or daily limits, and a clear no-storage policy.
The team aims for 10,000 monthly sessions to meet Ezoic's acceptance threshold for publishers, which would generate around $200–$500 per month with their current RPM of $8–$15 per session. Scaling to 50,000 sessions would make Mediavine a more attractive option to boost revenue.
DailyProTasks, the new product, is a collection of 5 free image/file tools hosted without signups, watermarks or file storage. The tools include a background remover, image upscaler, compressor, converter, and PDF generator. Development took three weeks, with infrastructure costs around $7 per month.
Currently, there are only 40 users, most of whom are friends and personal testers. The backend is built with FastAPI and Python, while the frontend uses vanilla HTML/CSS/JS and a PWA service worker. Key lessons learned include avoiding Windows Server on EC2 due to memory constraints, optimizing RAM usage for ML models, and troubleshooting SSL configuration on Windows.
The monetization plan involves reaching 10,000 sessions first, then applying for Ezoic and potentially moving to Mediavine for increased ad revenue. The author hopes to achieve $500–$1,500 per month in revenue by month 6.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.