Case study
Jobblish
A freelance marketplace where the chat feels instant.
- Role
- Senior Software Engineer and Team Lead
- Zee Palm
- When
- Zee Palm era
- Where
- Client delivery
The problem
What made this hard
A marketplace lives or dies on the conversation between client and freelancer. Jobblish put real-time chat and payments at the center: post work, talk it through, pay, all inside the app.
What shipped
Built, wired, released
Real-time messaging on Stream. Channels per job and presence built in, instead of hand-rolling sockets and read receipts.
Stripe wired to the hiring flow. Agreeing on a price and paying it are the same motion.
The marketplace core. Job posting, discovery and profiles, built as a cross-platform Flutter app.
Architecture
How it holds together
L0Stream handles the hard parts of chat (delivery, presence, history); the app composes its SDK rather than rebuilding it.
L1Stripe on the payment path with Firebase as the backend of record.
Screens
From the actual build
Every frame is a real capture unless labeled otherwise. Tap any screen to open it.
Outcome
Where it landed
- Delivered as part of the Zee Palm production run of 10+ Flutter apps.
- The screens above are frames from the product's Figma file, labeled as design frames; no captures survived, and none are faked.