{"data":{"items":[{"id":"b8439338-b240-4cc1-8860-082989f02ebb","excerpt":"I built and launched an AI weather app in 3 weeks using “vibe-coding” — My latest “vibe-coding” project: **Iso Weather**.\n\nIn December, I built and launched a fully native iOS app in about three weeks. After launch I kind of fell in love with the project and spent way more time polishing it than originally planned.\n\nWh","url":"https://www.reddit.com/r/VibeCodeDevs/comments/1r0y4qa/i_built_and_launched_an_ai_weather_app_in_3_weeks/","role":"request","weight":1.1503334,"occurredAt":"2026-02-10T11:20:50.000Z","sourceKey":"reddit","sourceName":"Reddit","credibility":0.62,"venue":"VibeCodeDevs","intent":"problem_report","painScore":0.36,"sentiment":0.45454547,"confidence":0.84583336,"matchedPatterns":["manual_process","urgent"],"statement":"Overall, I shipped this **much faster** than if I had coded everything manually.","title":"I built and launched an AI weather app in 3 weeks using “vibe-coding”","body":"My latest “vibe-coding” project: **Iso Weather**.\n\nIn December, I built and launched a fully native iOS app in about three weeks. After launch I kind of fell in love with the project and spent way more time polishing it than originally planned.\n\nWhat surprised me the most was how fast you can move now without writing that much code yourself. To be fair, I have a pretty long background in app development, and I picked a stack I know well: **Swift + Firebase + TypeScript** for the backend. I also built a small React admin panel, which is definitely not my strongest area, so AI saved me a lot of time there.\n\n**Main AI tools I used:**\n\n* OpenAI Codex CLI\n* Claude CLI\n\n**Other tooling:**\n\n* Xcode\n* Tower for macOS for mac\n* Github for CI/CD and code repo\n* Fastlane for automating App Store metadata uploads\n* [Shots.so](http://shots.so/) for promotional screenshots\n* [Appscreens.com](http://appscreens.com/) för App Store screenshots\n* RevenueCat for subscriptions and Paywalls\n\nWith a background in iOS development I could review all code, but with a deadline of Christmas and a lot of code produced I mainly just reviewed the resulting experience and did a shallow review of the code being produced. \n\nThe app itself is an **AI-integrated weather app** that generates a small **isometric city scene** based on the location, weather, temperature, season, and time of day.\n\nThe images are generated on demand and then cached in the backend (Firebase), so they don’t need to be regenerated every time. Each city can have around 100 variations. Since each generated image costs about €0.10, I had to keep a close eye on the economics. That made a subscription model necessary.\n\n**For monetization I used:**\n\n* iOS subscriptions\n* RevenueCat for paywalls and A/B testing\n\nThe app is live on the App Store. It’s gotten some nice traction in Sweden after a few LinkedIn posts. I’ve now launched it across Europe and globally, but downloads are still pretty modest outside my home market.\n\nNext step is marketing. I just started experimenting with App Store Ads. I also launched a small website (AI-generated, of course) and started publishing AI-written SEO articles. Too early to say how that will perform.\n\nAfter writing in other Reddit groups the main concern seems to be the pricing. At my original pricing at $50 per and only a weekly alternative priced quite high (to direct users towards the yearly plan) a lot of users complained. Seems reasonable, but at the same time I needed to covert the backend AI costs. I after this feedback lowered the variations from 100 to 60 per city and lowered from 2 to 1 custom city generation per month. This allowed me to lower pricing per year to $25. I also added a monthly plan for $3.99. Hope this pricing would be more acceptable (still high for a weather app I know, but can't go lower that my backend costs).\n\nOverall, I shipped this **much faster** than if I had coded everything manually. Especially the React webb admin, which would have taken me significantly longer on my own. An experienced developer still helps a lot though—I could solve the hardest parts myself and rarely got stuck for long.\n\nBut regardless, it felt like a huge creativity boost. In just a few weeks, I was able to launch a fairly advanced service:\n\n* Polished native app\n* Backend\n* Authentication\n* Payments\n* Webb admin system\n\nI have a long background in iOS developer so I would probably be able to build it from scratch. But it would have taken a lot longer and probably have more issues.\n\nCurious how others are experiencing this new “AI-assisted” development style. Is it speeding you up as much as it did for me? \n\nAlso, a review on App Store if you tried it would be appreciated!  \n[https://apps.apple.com/se/app/iso-weather-isometric-with-ai/id6756013340](https://apps.apple.com/se/app/iso-weather-isometric-with-ai/id6756013340)","offTopic":true},{"id":"ef280df2-a0c9-41b8-b22c-a6a47769d061","excerpt":"I built and launched an AI weather app in 3 weeks using “vibe-coding” — My latest “vibe-coding” project: **Iso Weather**.\n\nIn December, I built and launched a fully native iOS app in about three weeks. After launch I kind of fell in love with the project and spent way more time polishing it than originally planned.\n\nWh","url":"https://www.reddit.com/r/vibecoding/comments/1r0xx18/i_built_and_launched_an_ai_weather_app_in_3_weeks/","role":"request","weight":1.1503334,"occurredAt":"2026-02-10T11:08:46.000Z","sourceKey":"reddit","sourceName":"Reddit","credibility":0.62,"venue":"vibecoding","intent":"problem_report","painScore":0.36,"sentiment":0.45454547,"confidence":0.84583336,"matchedPatterns":["manual_process","urgent"],"statement":"Overall, I shipped this **much faster** than if I had coded everything manually.","title":"I built and launched an AI weather app in 3 weeks using “vibe-coding”","body":"My latest “vibe-coding” project: **Iso Weather**.\n\nIn December, I built and launched a fully native iOS app in about three weeks. After launch I kind of fell in love with the project and spent way more time polishing it than originally planned.\n\nWhat surprised me the most was how fast you can move now without writing that much code yourself. To be fair, I have a pretty long background in app development, and I picked a stack I know well: **Swift + Firebase + TypeScript** for the backend. I also built a small React admin panel, which is definitely not my strongest area, so AI saved me a lot of time there.\n\n**Main AI tools I used:**\n\n* OpenAI Codex CLI\n* Claude CLI\n\n**Other tooling:**\n\n* Xcode\n* Tower for macOS for mac\n* Github for CI/CD and code repo\n* Fastlane for automating App Store metadata uploads\n* [Shots.so](http://Shots.so) for promotional screenshots\n* [Appscreens.com](http://Appscreens.com) för App Store screenshots\n* RevenueCat for subscriptions and Paywalls\n\nWith a background in iOS development I could review all code, but with a deadline of Christmas and a lot of code produced I mainly just reviewed the resulting experience and did a shallow review of the code being produced.\n\nThe app itself is an **AI-integrated weather app** that generates a small **isometric city scene** based on the location, weather, temperature, season, and time of day.\n\nThe images are generated on demand and then cached in the backend (Firebase), so they don’t need to be regenerated every time. Each city can have around 100 variations. Since each generated image costs about €0.10, I had to keep a close eye on the economics. That made a subscription model necessary.\n\n**For monetization I used:**\n\n* iOS subscriptions\n* RevenueCat for paywalls and A/B testing\n\nThe app is live on the App Store. It’s gotten some nice traction in Sweden after a few LinkedIn posts. I’ve now launched it across Europe and globally, but downloads are still pretty modest outside my home market.\n\nNext step is marketing. I just started experimenting with App Store Ads. I also launched a small website (AI-generated, of course) and started publishing AI-written SEO articles. Too early to say how that will perform.\n\nAfter writing in other Reddit groups the main concern seems to be the pricing. At my original pricing at $50 per and only a weekly alternative priced quite high (to direct users towards the yearly plan) a lot of users complained. Seems reasonable, but at the same time I needed to covert the backend AI costs. I after this feedback lowered the variations from 100 to 60 per city and lowered from 2 to 1 custom city generation per month. This allowed me to lower pricing per year to $25. I also added a monthly plan for $3.99. Hope this pricing would be more acceptable (still high for a weather app I know, but can't go lower that my backend costs).\n\nOverall, I shipped this **much faster** than if I had coded everything manually. Especially the React webb admin, which would have taken me significantly longer on my own. An experienced developer still helps a lot though—I could solve the hardest parts myself and rarely got stuck for long.\n\nBut regardless, it felt like a huge creativity boost. In just a few weeks, I was able to launch a fairly advanced service:\n\n* Polished native app\n* Backend\n* Authentication\n* Payments\n* Webb admin system\n\nI have a long background in iOS developer so I would probably be able to build it from scratch. But it would have taken a lot longer and probably have more issues.\n\nCurious how others are experiencing this new “AI-assisted” development style. Is it speeding you up as much as it did for me? \n\nAlso, a review on App Store if you tried it would be appreciated!\n\n  \nDownload link: [https://apps.apple.com/se/app/iso-weather-isometric-with-ai/id6756013340](https://apps.apple.com/se/app/iso-weather-isometric-with-ai/id6756013340)","offTopic":true},{"id":"8225f139-4af2-43a8-ab40-7988f3c8512c","excerpt":"My latest SaaS-app: Iso Weather. Built and launched in 3 weeks! — In December, I built and launched a fully native iOS app in about three weeks. After launch I kind of fell in love with the project and spent way more time polishing it than originally planned.\n\nWhat surprised me the most was how fast you can move now wi","url":"https://www.reddit.com/r/microsaas/comments/1r0y78z/my_latest_saasapp_iso_weather_built_and_launched/","role":"request","weight":1.1360464,"occurredAt":"2026-02-10T11:24:48.000Z","sourceKey":"reddit","sourceName":"Reddit","credibility":0.62,"venue":"microsaas","intent":"problem_report","painScore":0.36,"sentiment":0.45454547,"confidence":0.8353282,"matchedPatterns":["manual_process","urgent","product:openai"],"statement":"Overall, I shipped this **much faster** than if I had coded everything manually.","title":"My latest SaaS-app: Iso Weather. Built and launched in 3 weeks!","body":"In December, I built and launched a fully native iOS app in about three weeks. After launch I kind of fell in love with the project and spent way more time polishing it than originally planned.\n\nWhat surprised me the most was how fast you can move now without writing that much code yourself. To be fair, I have a pretty long background in app development, and I picked a stack I know well: **Swift + Firebase + TypeScript** for the backend. I also built a small React admin panel, which is definitely not my strongest area, so AI saved me a lot of time there.\n\n**Main AI tools I used:**\n\n* OpenAI Codex CLI\n* Claude CLI\n\n**Other tooling:**\n\n* Xcode\n* Tower for macOS for mac\n* Github for CI/CD and code repo\n* Fastlane for automating App Store metadata uploads\n* [Shots.so](http://shots.so/) for promotional screenshots\n* [Appscreens.com](http://appscreens.com/) för App Store screenshots\n* RevenueCat for subscriptions and Paywalls\n\nWith a background in iOS development I could review all code, but with a deadline of Christmas and a lot of code produced I mainly just reviewed the resulting experience and did a shallow review of the code being produced. \n\nThe app itself is an **AI-integrated weather app** that generates a small **isometric city scene** based on the location, weather, temperature, season, and time of day.\n\nThe images are generated on demand and then cached in the backend (Firebase), so they don’t need to be regenerated every time. Each city can have around 100 variations. Since each generated image costs about €0.10, I had to keep a close eye on the economics. That made a subscription model necessary.\n\n**For monetization I used:**\n\n* iOS subscriptions\n* RevenueCat for paywalls and A/B testing\n\nThe app is live on the App Store. It’s gotten some nice traction in Sweden after a few LinkedIn posts. I’ve now launched it across Europe and globally, but downloads are still pretty modest outside my home market.\n\nNext step is marketing. I just started experimenting with App Store Ads. I also launched a small website (AI-generated, of course) and started publishing AI-written SEO articles. Too early to say how that will perform.\n\nAfter writing in other Reddit groups the main concern seems to be the pricing. At my original pricing at $50 per and only a weekly alternative priced quite high (to direct users towards the yearly plan) a lot of users complained. Seems reasonable, but at the same time I needed to covert the backend AI costs. I after this feedback lowered the variations from 100 to 60 per city and lowered from 2 to 1 custom city generation per month. This allowed me to lower pricing per year to $25. I also added a monthly plan for $3.99. Hope this pricing would be more acceptable (still high for a weather app I know, but can't go lower that my backend costs).\n\nOverall, I shipped this **much faster** than if I had coded everything manually. Especially the React webb admin, which would have taken me significantly longer on my own. An experienced developer still helps a lot though—I could solve the hardest parts myself and rarely got stuck for long.\n\nBut regardless, it felt like a huge creativity boost. In just a few weeks, I was able to launch a fairly advanced service:\n\n* Polished native app\n* Backend\n* Authentication\n* Payments\n* Webb admin system\n\nI have a long background in iOS developer so I would probably be able to build it from scratch. But it would have taken a lot longer and probably have more issues.\n\nCurious how others are experiencing this new “AI-assisted” development style. Is it speeding you up as much as it did for me? \n\nAlso, a review on App Store if you tried it would be appreciated!\n\n","offTopic":true},{"id":"767fb109-c7d8-48f1-9792-53b8851da49e","excerpt":"I built and launched an AI weather app in 3 weeks using “vibe-coding” — My latest “vibe-coding” project: **Iso Weather**.\n\nIn December, I built and launched a fully native iOS app in about three weeks. After launch I kind of fell in love with the project and spent way more time polishing it than originally planned.\n\nWh","url":"https://www.reddit.com/r/iOSAppsMarketing/comments/1r0xr12/i_built_and_launched_an_ai_weather_app_in_3_weeks/","role":"request","weight":0.9916667,"occurredAt":"2026-02-10T10:59:17.000Z","sourceKey":"reddit","sourceName":"Reddit","credibility":0.62,"venue":"iOSAppsMarketing","intent":"problem_report","painScore":0.36,"sentiment":0.45454547,"confidence":0.7291667,"matchedPatterns":["manual_process"],"statement":"Overall, I shipped this **much faster** than if I had coded everything manually.","title":"I built and launched an AI weather app in 3 weeks using “vibe-coding”","body":"My latest “vibe-coding” project: **Iso Weather**.\n\nIn December, I built and launched a fully native iOS app in about three weeks. After launch I kind of fell in love with the project and spent way more time polishing it than originally planned.\n\nWhat surprised me the most was how fast you can move now without writing that much code yourself. To be fair, I have a pretty long background in app development, and I picked a stack I know well: **Swift + Firebase + TypeScript** for the backend. I also built a small React admin panel, which is definitely not my strongest area, so AI saved me a lot of time there.\n\n**Main tools I used:**\n\n* OpenAI Codex CLI\n* Claude CLI\n\nThe app itself is an **AI-integrated weather app** that generates a small **isometric city scene** based on the location, weather, temperature, season, and time of day.\n\nThe images are generated on demand and then cached in the backend (Firebase), so they don’t need to be regenerated every time. Each city can have around 100 variations. Since each generated image costs about €0.10, I had to keep a close eye on the economics. That made a subscription model necessary.\n\n**For monetization I used:**\n\n* iOS subscriptions\n* RevenueCat for paywalls and A/B testing\n\nThe app is live on the App Store. It’s gotten some nice traction in Sweden after a few LinkedIn posts. I’ve now launched it across Europe and globally, but downloads are still pretty modest outside my home market.\n\nNext step is marketing. I just started experimenting with App Store Ads. I also launched a small website (AI-generated, of course) and started publishing AI-written SEO articles. Too early to say how that will perform.\n\nAfter writing in other Reddit groups the main concern seems to be the pricing. At my original pricing at $50 per and only a weekly alternative priced quite high (to direct users towards the yearly plan) a lot of users complained. Seems reasonable, but at the same time I needed to covert the backend AI costs. I after this feedback lowered the variations from 100 to 60 per city and lowered from 2 to 1 custom city generation per month. This allowed me to lower pricing per year to $25. I also added a monthly plan for $3.99.\n\nOverall, I shipped this **much faster** than if I had coded everything manually. Especially the React webb admin, which would have taken me significantly longer on my own. An experienced developer still helps a lot though—I could solve the hardest parts myself and rarely got stuck for long.\n\nBut regardless, it felt like a huge creativity boost. In just a few weeks, I was able to launch a fairly advanced service:\n\n* Polished native app\n* Backend\n* Authentication\n* Payments\n* Webb admin system\n\nI have a long background in iOS developer so I would probably be able to build it from scratch. But it would have taken a lot longer and probably have more issues.\n\nCurious how others are experiencing this new “AI-assisted” development style. Is it speeding you up as much as it did for me? \n\nAlso, a review on App Store if you tried it would be appreciated! Iso Weather is the name.\n\n[https://apps.apple.com/se/app/iso-weather-isometric-with-ai/id6756013340](https://apps.apple.com/se/app/iso-weather-isometric-with-ai/id6756013340)","offTopic":true}],"breakdown":[{"sourceKey":"reddit","sourceName":"Reddit","count":4}],"total":4}}