Questions an interviewer could realistically ask about this project, drawn from
../learning/. Gees Arise is an accountability app: circles of friends set tasks,
submit proof, audit each other, and take penalties — built on Next.js 16 + Supabase (Postgres, RLS,
Storage, RPC functions, cron).

This is the strongest project in the vault for database and security questions. The
09-sys-design.md notes contain a genuinely unusual amount of
real Postgres/RLS depth: recursive policy traps, SECURITY DEFINER used correctly, partial unique
indexes, triggers as cross-feature glue, and several bugs that no type system could have caught.

How to use this

  • Answer out loud, from memory, before reading the hint.
  • Strong answer covers = the checklist a good answer hits, not a script.
  • [Beginner] / [Intermediate] / [Advanced] = how much project context the question assumes.
  • 🔥 = most likely to be asked. 🔥🔥 = build your prep around it.

Files

FileCovers
01-postgres-rls-and-security.mdRLS, the recursion trap, SECURITY DEFINER, RPC vs policy, storage policies, signed URLs
02-data-modeling-and-invariants.mdRule vs log tables, state machines, constraints over app code, triggers, partial unique indexes, fan-out
03-nextjs-and-frontend.mdRoute groups, Server Components/Actions, the upload-limit story, optimistic UI, stale layouts, Playwright
04-bugs-and-story.mdThe silent bugs, git workflow, trade-offs, behavioural

Before anything else: the 60-second pitch

Gees Arise is a group accountability app — you and a circle of friends commit to tasks, submit
proof, and the circle audits each other; missed or flagged tasks become penalties, resolved by
vote. It’s Next.js 16 on Supabase, and the interesting engineering is almost entirely in the
database. Supabase auto-exposes a REST API over every public table with a key that ships in the
browser bundle, so the access rules have to live in Postgres, not in my route handlers —
otherwise “only circle members can see this” is only true because my app happens to check it.
That pushed a lot of design into RLS policies, SECURITY DEFINER RPC functions for atomic
multi-table actions, database constraints for the abuse guarantees, and triggers so one feature
can react to another’s writes without either knowing the other exists.

The second half is the pitch. “The security boundary is the database, because the client key is
public” is a one-sentence demonstration that you understand the platform rather than just using it.

↑ Back to Projects — Learning Logs