Now hiring
Disincentive is a small team shipping an iOS commitment-contract app that's weeks from the App Store. We use AI tooling to move faster than the size of our team should let us. We need people who do the same.
What's been built
The codebase you'd inherit on day one. Imagine what we ship with two of you.
The craft
A real bundle on the left. A real piece of the streak engine on the right. No mystery box. If this kind of code looks like home, you're our person.
$ cd disincentive-app && yarn ios # Resolving deps, compiling TS, building Spirit... → Compiling 4,847 lines of TypeScript → Resolving 247 dependencies → Building Spirit (754 lines, 140 variants) → Patching Stripe webhook handler → Pushing migration 0008_support_messages.sql → Running test suite (43 tests, 12 skipped) → Creating archive (iPhone 15 Pro) ✓ Build ready in 47.2s. # Now do this 200 more times before submission.
src/streak/engine.ts// derives the current tier from a rule's check-ins. // time-zone aware, grace-day aware, idempotent. export function computeStreakState( rule: Rule, checkIns: CheckIn[], now: Date = new Date(), ): StreakState { const days = countKeptDays(checkIns, now); const tier = tierFor(days); const graceMax = graceQuota(tier); return { tier, days, canUseGrace: graceUsed(checkIns) < graceMax, nextTier: nextTierAfter(tier), }; } const TIERS = [ { name: "Quiet", from: 0 }, { name: "Steady", from: 7 }, { name: "Forged", from: 30 }, { name: "Sworn", from: 90 }, { name: "Kept", from: 365 }, ] as const;
What you'd own
We're past prototype. Disincentive bundles cleanly, runs on real devices, and hits Supabase / Stripe end-to-end in a sandbox. What's left is the part that matters most: making it survive contact with real users, real money, and the App Store review process.
Concrete things on the next ninety days of the roadmap: the Stripe webhook for forfeit capture in production. Push notification scheduling across time zones with grace days and partial-completion logic. The Spirit growth pipeline — 140 SVG variants tied to streak state and rendered without jank. Edge Functions for daily streak calculations, run via Supabase cron. App Store §5.3 compliance for a financial-commitment device that isn't gambling, isn't a lottery, isn't a contest, but lives adjacent to all three. Real product engineering, not framework chasing.
AI tooling is part of the day, not a buzzword. We use Claude Code and Cursor in the loop continuously — for migrations, for refactors, for App Store copy, for legal drafting, for thinking. We expect anyone joining to bring their own sharpened tools and to make the team faster, not slower. If you've shipped real code in the last six months that you wrote alongside an LLM, you'll feel at home.
The stack
No buzzword bingo. No "we're considering rewriting in Rust." Just what's running, today, in production.
Open roles
Both are senior contracts to start, with a path to founding-team equity if it clicks. We don't post jobs we don't have. Pay is real, comp is transparent.
Own the iOS app end to end. You'd inherit a production-clean Expo / React Native codebase and ship it to the App Store with us, then iterate on the v2 surface area as users come in.
Build the agent and AI tooling that powers our internal workflow and (eventually) user-facing features. You ship code daily alongside Claude / Cursor and have strong opinions about where humans should stay in the loop.
The team
Small. Senior. No standups. We talk when we have something to say. The cards below get more real every quarter.
Built Disincentive end to end — mobile, backend, payments, design system, App Store strategy. Comes from a product-engineering background. Spends most of his day in Claude Code, Cursor, and the iOS Simulator.
The person who'll own the iOS app from launch through 1.0. Senior. Ships things. Cares about the streak counter being correct on a Friday at 11:58pm just as much as on a Tuesday at noon.
The person who runs our AI tooling like an instrument. Production-grade Anthropic API work, internal agent flows, and a strong stomach for figuring out what AI in the product should look like.
The deal
Pay is real. Contract rates above are honest market rates for senior engineers. We don't underpay because we're "early." If we can't pay you what your time is worth, we'll say so before you start.
Equity is on the table. If you ship with us through launch and the trajectory is right, founding-team equity is part of the conversation. We'll be specific about strike, vest, and dilution in writing — no hand-waving.
Hours are sane. No 60-hour weeks. No "we're a family" emails. The deal is the deal: ship good work on agreed timelines, communicate when something breaks, and we'll defend your time like we defend our own.
AI is part of the day. We expect you to ship with Claude / Cursor in the loop continuously. We pay for the tools. We expect you to have opinions about where they help and where they don't.
Common questions
One async writeup (a paragraph in your application email). One 60-minute conversation about a real engineering problem we're working on right now. One paid trial — typically 8–16 hours of real product work, billed at your rate. If everyone's still nodding by the end of the trial, we sign a contract. No leetcode, no take-home tests, no panel of strangers.
Helpful, not required. Senior engineers learn a stack quickly. We care more that you've shipped real things and can ramp up on whatever's in front of you. If you've never touched Expo or Supabase but you've built production iOS apps, you'll be fine. If you've never touched TypeScript, that's a harder sell.
Forty hours a week, or whatever you and we agree on in writing. We don't care about timezone overlap as long as you ship and you're reachable for one or two stand-ins per week. We don't track keyboards. The deal is the deal: agreed scope on agreed dates.
Yes to both. Most of us would rather have a great engineer for 20 hours a week than an okay one for 40. And contract-only is fine — equity is a separate conversation we'll only have if you want to have it.
You ship code with Claude or Cursor open, basically all the time. You write prompts, you read diffs, you push back when the model's wrong, and you commit fast when it's right. We're not looking for "AI experts" — we're looking for engineers who already work this way and can show us where the model helps and where it gets in the way.
If you ship through launch and we're growing, founding-team equity is real. Specific percentages, vest schedule, strike price, and dilution math get put in writing before you commit. We don't dangle equity as a substitute for cash — cash comes first, and equity is on top of a market rate.
Because we wanted you to know the difference between a marketing page and a recruiting page without anyone having to say it.
Email us a paragraph about something you shipped recently and what you wrote next to an LLM to get there. No résumé needed. We'll read every one and reply within 48 hours, even if it's a no.
hello@disincentive.app →