developer coding user risk: medium

Expo Supabase AI Generation Architect

Instructs the model to act as a Senior Expo + Supabase Architect and implement a cold-start safe architecture using Expo client, Supabase Postgres, Storage, Realtime, Edge Function…

Policy sensitive
Human review

PROMPT

Act as a Senior Expo + Supabase Architect.

Implement a “cold-start safe” architecture using:
- Expo (React Native) client
- Supabase Postgres + Storage + Realtime
- Supabase Edge Functions ONLY for lightweight gating + job enqueue
- A separate Worker service for heavy AI generation and storage writes

Deliver:
1) Database schema (SQL migrations) for: jobs, generations, entitlements (credits/is_paid), including indexes and RLS notes
2) Edge Functions:
   - ping (HEAD/GET)
   - enqueue_generation (validate auth, check is_paid/credits, create job, return jobId)
   - get_job_status (light read)
   Keep imports minimal; no heavy SDKs.
3) Expo client flow:
   - non-blocking warm ping on app start
   - Generate button uses optimistic UI + placeholder
   - subscribe to job updates via Realtime or implement polling fallback
   - final generation replaces placeholder in gallery list
4) Worker responsibilities (describe interface and minimal endpoints/logic, do not overbuild):
   - fetch queued jobs
   - run AI generation
   - upload to storage
   - update jobs + insert generations
   - retry policy and idempotency

Constraints:
- Do NOT block app launch on any Edge call
- Do NOT run AI calls inside Edge Functions
- Ensure failed jobs still create a generation record with original input visible
- Keep the solution production-friendly but minimal

Output must be structured as:
A) Architecture summary
B) Migrations (SQL)
C) Edge function file structure + key code blocks
D) Expo integration notes + key code blocks
E) Worker outline + pseudo-code

ROLES & RULES

Role assignments

Act as a Senior Expo + Supabase Architect.

Keep imports minimal; no heavy SDKs.
Do NOT block app launch on any Edge call
Do NOT run AI calls inside Edge Functions
Ensure failed jobs still create a generation record with original input visible
Keep the solution production-friendly but minimal

EXPECTED OUTPUT

Format

structured_report

Schema

markdown_sections · Architecture summary, Migrations (SQL), Edge function file structure + key code blocks, Expo integration notes + key code blocks, Worker outline + pseudo-code

Constraints

structured as A) Architecture summary
B) Migrations (SQL)
C) Edge function file structure + key code blocks
D) Expo integration notes + key code blocks
E) Worker outline + pseudo-code

SUCCESS CRITERIA

Deliver database schema (SQL migrations) for jobs, generations, entitlements including indexes and RLS notes
Provide Edge Functions: ping, enqueue_generation, get_job_status
Describe Expo client flow: non-blocking warm ping, optimistic UI, subscribe to job updates, final generation in gallery
Outline Worker responsibilities: fetch queued jobs, run AI generation, upload to storage, update jobs and insert generations, retry policy and idempotency

FAILURE MODES

Blocking app launch on Edge calls
Running AI calls inside Edge Functions
Not creating generation records for failed jobs
Overbuilding Worker or using heavy SDKs
Ignoring cold-start safety or production constraints

CAVEATS

Missing context

Specific AI service or model for Worker generation (e.g., OpenAI, Stability AI)
Type of AI generation output (e.g., image, text, video)
Pre-existing app structure or gallery component details

QUALITY

OVERALL: 0.85
CLARITY: 0.95
SPECIFICITY: 0.95
REUSABILITY: 0.25
COMPLETENESS: 0.90

IMPROVEMENT SUGGESTIONS

Introduce placeholders for AI service, generation type, and custom schema fields to boost reusability
Explicitly define data shapes (e.g., JSON examples for job payload) in constraints
Add success criteria like expected latency or error rates for production-friendliness

USAGE

Copy the prompt above and paste it into your AI of choice — Claude, ChatGPT, Gemini, or anywhere else you're working. Replace any placeholder sections with your own context, then ask for the output.