Build · Create · Ops & Automation
Job requisition spec
A complete requisition: responsibilities with success metrics, a must-have skills list capped to avoid the unicorn trap, a valid comp range, and a clean biased-language screen.
You receive: A structured requisition (JSON) that passes the completeness, skill-cap, comp, and biased-language checks.
Part of Lead Team
Example
A sample of what this play produces. Your result is generated for your inputs.
Role title
Founding Engineer, Play Catalog and Judge
Location mode
remote
Role summary field
STUD is a living system for a venture: it knows the business, does the work, and learns from every result. The catalog today is 265 plays across 14 playbooks, each settled by a machine acceptance check before a deliverable counts. You are the first engineering hire alongside the founder. You own the play catalog and the judge harness: authoring new settleable plays, hardening the validator and oracle layer, and keeping the deploy path (Next.js on Vercel, Neon Postgres, the cred-free CI gate) production-clean. Concierge fulfillment currently caps at about 30 plays per week on one founder-operator; your work raises that ceiling.
Company why field
AI can produce real knowledge work, but getting that work out of it is still manual craft: you have to know what to ask, re-explain your business every session, and carry the output somewhere by hand. STUD replaces the blank prompt with ready-made plays that fill their intake from workspace facts, so a venture's knowledge compounds instead of going stale. Pre-seed, pre-revenue soft launch, founded by Dan Schmitz: you would be shaping the core product, not maintaining someone else's.
Growth path field
Start as the sole engineer owning catalog and judge reliability. As paid fulfillment opens, grow into the technical lead for the settlement platform (judge worker, marketplace supply integrations), with a path to founding CTO scope as the team forms around you.
Responsibilities
| Duty | Success metric |
|---|---|
| Author and integrate new settleable plays into the live catalog (validator, intake schema, sample deliverable) using the existing multi-agent integration harness. | Catalog grows by at least 20 net new plays per month, every one green through the full cred-free gate on first ship. |
| Harden the judge harness: process isolation, held-out case coverage, and oracle correctness audits for calculator plays. | Zero judge escapes (an accepted deliverable later found failing its own published contract) per quarter. |
| Own the deploy path end to end: schema migrations, catalog sync on deploy, and the CI gate that blocks regressions. | Production deploys stay under 15 minutes commit-to-live with no manual prod steps and no failed-migration incidents. |
| Instrument fulfillment throughput so play runs move off the founder-operator concierge path. | Automated run capacity exceeds the current 30 plays per week concierge ceiling within two quarters. |
Must have skills
| Name | Type |
|---|---|
| TypeScript and Next.js (App Router) in production | technical |
| Python for test harnesses and validation tooling | technical |
| SQL schema design and migration discipline (Postgres) | technical |
| CI/CD pipeline ownership (GitHub Actions or equivalent) | technical |
| Written communication clear enough to double as documentation | soft |
| Comfort operating solo with founder-level ambiguity | soft |
Nice to have skills
| Name | Type |
|---|---|
| LLM agent orchestration or eval-harness experience | technical |
| Marketplace or usage-billing product experience | technical |
| Prior early-stage (first-five-employees) startup tenure | soft |
Comp range
- Min
- 140000
- Max
- 180000
- Currency
- USD
- Note
- Illustrative range for planning: STUD is pre-seed and pre-revenue, and this requisition is the planning spec, not a committed offer. Equity component sized at offer stage.
Opens soon
STUD opens runs for everyone soon, on every play at once.
Get early access to STUD the day it goes live.