Software Engineer - Product, Full Stack

About the role

We're hiring a product engineer to build across our verification platform: Python services, React frontends, and the data model underneath. You'll own whole slices of the product instead of tickets handed down in pieces, and work close enough to real customers to feel what's awkward about our product and fix it. Unlike most roles here, this one comes with senior engineers next to you to learn from. We care more about how fast you climb than where you're starting.

Responsibilities

• Build and ship features end to end: our Python services, the React and TypeScript frontends, and the schema underneath.
• Build the integration surface customers actually meet: session initialization, the embedded SDK, API keys, and the path from signing up to a first successful verification.
• Work on the signal and trust surface: the flags, signals, and scoring that turn what we observed into a verdict an operator can act on, plus the evidence trail that makes that verdict explicable.
• Work directly on pilot customer integrations, close enough to real usage to feel what's awkward and go fix it.
• Write the tests for what you ship. Building the coverage for your own work beats inheriting someone else's.
• Take and give code review. On a team this size, review is how quality holds.

The charter

• HumanCheck distribution and outcomes: The round trip: a customer decides who needs verifying, we get a check to them, and the results land back in the system they actually work in. Currently our most urgent product track.
• Integration model: How a company integrates us: which principal, which keys, which session-init path, and what the SDK and tokens actually promise.
• Roles and permissions: A real privilege model across both the internal and customer-facing dashboards. Genuinely a whole slice of product, not a settings screen.
• Reporting and insights: Dashboards, aggregates, and exports over the signals we derive.
• Product demo experience: A public, no-login demo of the verification flow, so prospects can understand uniqueness without a sales call.
• UI Design: MVP screens shipped ahead of the designed experience; bringing them up to it, one screen at a time.

Our stack

• Backend: Python 3.14, FastAPI, Pydantic v2, SQLAlchemy 2 (async) with asyncpg, Alembic, Strawberry GraphQL.
• Frontend: React + TypeScript, Vite, TanStack Router / Query / Form, Tailwind, Radix UI.
• Data: Cloud SQL Postgres 18, Firestore, Valkey, BigQuery, Pub/Sub.
• Cloud: GCP, Cloud Run. The whole stack runs locally as one docker compose project with emulators.
• Integrations you'd touch: Clerk for partner auth, AWS for specific parts of our architecture, Twilio for SMS, Postmark for distribution email, PostHog for product analytics.
• Tooling: uv, ruff, pyright, pytest, pre-commit.

You'll write both backend and frontend. We'll help you get good at the half you're weaker on.

Requirements

• You've shipped software that real people used: a job, an internship, a co-op, or something you built yourself. We're deliberately not setting a years bar, and we screen on evidence instead of tenure.
• Strong in Python or TypeScript, and genuinely willing on the other. You'll write both.
• You can reason about a relational schema, and about what it takes to change one that already has production data in it.
• Solid fundamentals: HTTP, relational data, API design. We care that you can reason from them, not recite them.
• Evidence you can debug. Our interview asks about a bug you couldn't reproduce, because that's the skill that compounds.
• Comfort somewhere small enough that "not my area" isn't an available answer.
• Fluency with AI tooling, and the judgment to check it. Coding agents and AI-assisted review are part of our normal workflow, and we expect you to be good with them while still owning the correctness of what you ship.

Nice to have

• Identity, fraud, trust and safety, or anti-abuse exposure.
• React with TanStack Router or Query, or a comparable modern data-fetching stack.
• GraphQL.
• Any experience with GCP, Docker, or CI.

How we work

Small team, high autonomy, low ceremony. You'd have a real stake in technical direction instead of inheriting someone else's five-year-old decisions. Most of what we're building is close to greenfield.

We run on Linear for execution, Figma for design, Notion for anything durable, and GitHub with agentic code review in the loop. We care a lot about developer experience. If a workflow gets in the way of shipping, we fix it.

Why VerifyYou

The internet has evolved into critical infrastructure, but its trust model hasn't kept pace. AI has made synthetic participation trivial, and legacy solutions like KYC, CAPTCHAs, and phone verification are either fragile or invasive. We're building verification infrastructure that proves humanness while keeping personal data private.

You'll own real surface area, ship to production fast, and work directly with the founders and senior technical leadership.

Pay

Competitive base, plus meaningful early-stage equity.

Schedule

Full-time, remote.