The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
SupportFlow AI is a customer support management system built with Python, PySide6, FastAPI and PostgreSQL.
It helps teams manage customer requests, assign priorities and responsibilities, track conversations, and generate AI-assisted response suggestions while keeping a human in control of the final reply.
I built the application structure, interface, backend integration, database workflows, API layer and automated testing.
How do you handle a new customer message arriving while someone is reviewing an AI draft? Does the suggestion stay tied to the conversation version it was generated from, or get refreshed before they send it?
Great question, the AI suggestion is generated from the conversation state at the moment the request is made and if a new customer message arrives while the draft is being reviewed, the current implementation doesn’t automatically version-lock or refresh that suggestion yet.
A...
That makes sense. Flagging the draft when a new message comes in would be useful, especially if someone’s halfway through editing it. Thanks for explaining where things stand!
The conversation-revision idea sounds useful. I would also check that revision on the server when accepting the send, so a message arriving between review and click cannot slip through. Mark the draft stale while keeping the human edits visible. Two useful regression cases: a...
Oh yeah!!! I especially like the idea of validating the conversation revision again server-side at send time, since checking only when the draft is generated would still leave a race condition between review and send.
Keeping the human-edited draft visible but marking it as stale...
Thanks, Ivan — that discussion prompted a small self-initiated failure lab. I reproduced stale approvals, concurrent duplicate submissions and a worker exiting after the mock receiver commits, before the sender records success.
The key distinction: a stable request ID needs to...
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Challenges
View all
Related posts
AI Chatbot App — 40% Engagement Lift for a Customer Support Tool
Project description:
The client needed an AI chatbot app built from the ground up — no existing design, no reference, just a goal: make it feel human, fast and easy to use. I designed the full experience in Figma — mapping user flows, building wireframes, then crafting a clean high-fidelity UI with a smooth onboarding screen, contextual quick-reply chips and a minimal visual language that keeps focus on the conversation. Delivered with handoff-ready specs and a reusable component library.
Result:
→ +40% user engagement
→ −25% response time
→ Improved customer satisfaction
Wowwwww I like the aesthetics
Intellica AI, a mobile AI workspace.
Chat, generation and document analysis behind one input field, with a prompt library so nobody starts from a blank screen.
Morrow — Modern Digital Experience
A modern, polished web experience built with a strong emphasis on visual identity, clean composition, and a seamless user journey.
I focused on transforming the design concept into a refined, responsive digital experience with carefully crafted layouts, typography, spacing, and interaction details. Every section was structured to feel intentional while maintaining a smooth and consistent experience across screen sizes.
What I worked on
Developed the complete responsive frontend experience
Translated the visual direction into a polished web interface
Built reusable and maintainable UI components
Focused on typography, spacing, composition, and visual hierarchy
Created responsive layouts for desktop and mobile
Refined interactions and UI details for a more premium experience
Optimized the overall interface for usability and consistency
Outcome: A clean, modern, and visually refined web experience designed to give the brand a strong digital presence and a memorable first impression.
