Case study
UpAlerts.
An AI assistant for freelancers — real-time job alerts, AI-written proposals and rate tooling — shipped to iOS and Android on Flutter and serving 10,000 daily active users.
- Engagement
- Product build and ongoing delivery
- Role
- One senior engineer — mobile architecture through release
- Platforms
- iOS, Android
- Outcome
- 10k daily active users, 40% lift in engagement
The problem
Freelancers win work by being early. A job posts, and the proposals that land in the first few minutes are the ones that get read. A daily digest is useless for that, and a polling app that checks every fifteen minutes is only slightly better. The product needed alerts that arrive while the job is still fresh, on a phone, without draining the battery that makes the phone useful.
What we built
A Flutter application for iOS and Android backed by Firebase. The core of it is the alert path: matching a new posting against a user’s skills and rates, then getting a push notification onto the device fast enough to matter. Everything else on the roadmap depends on that path staying quick and staying quiet — an alert system that cries wolf gets muted, and a muted alert system has no users.
Around it sit the tools that turn an alert into a submitted proposal: AI-drafted cover letters tailored to each posting, rate suggestions, profile optimisation and an analytics view so a freelancer can see which applications actually convert.
Delivery runs through CI/CD, because an app whose value is speed cannot wait a week for a fix to reach the stores.
The result
10,000 daily active users and a 40% increase in user engagement. The engagement number is the one worth reading: it says people came back, which for an alerts product is the only metric that distinguishes a useful tool from a notification people learn to ignore.
Stack
Flutter · Dart · Firebase · Push notifications · Real-time alerts · Analytics · CI/CD