Senior Back-End Developer

The role
You'll own backend workstreams end to end — from a one-page design to deployed, tested, documented features. Our backend is a modular NestJS/TypeScript service backed by PostgreSQL, with asynchronous AI pipelines built on Redis and BullMQ, realtime updates over WebSockets, and integrations with commercial LLM APIs and our own Python AI microservices. It runs on Google Cloud (Cloud Run) with CI/CD via GitHub Actions.

Concretely, you will:
Design and build new backend modules — clean boundaries, guards, DTOs, typed contracts — in a codebase that takes architecture seriously.
Build and harden async job pipelines: queue workers, retry semantics, deduplication, caching, realtime progress events.
Model data in PostgreSQL/Prisma under real-world constraints: transactions, row-level locking, encryption at rest for sensitive fields.
Integrate LLM-based features the production way: structured outputs, validation, fallbacks, cost/latency logging, and regression evals — treating the model as an unreliable external dependency to engineer around.
Work across the TypeScript↔Python boundary with our AI services (typed API contracts, versioned deployments).
Practice privacy engineering daily: we process special-category health data under GDPR Art. 9, so data minimization, audit logging, and safe error handling are part of every code review — not an afterthought.
Contribute to subscription and usage-metering logic (Stripe).

What we're looking for
Must have:
5+ years of backend engineering, with deep TypeScript/Node.js expertise; NestJS experience strongly preferred (or proven depth in a comparable DI/module framework and the appetite to go deep fast).
Strong relational-database skills: schema design, transactions, concurrency (you can reason about what happens when two workers write the same row — and pick the right locking strategy).
Real experience with event-driven/async systems: message queues, workers, pub/sub, idempotency, failure handling.
A security and privacy reflex. You notice the PII in the log statement without being told. Experience in a regulated domain (health, fintech, insurance) is a strong signal.
Experience integrating LLM APIs into production systems — or outstanding async-systems fundamentals and a clear interest in building this muscle.
You write tests and documentation as part of the work, not after it.

Nice to have:
Google Cloud (Cloud Run, IAM, secrets) and GitHub Actions.
Python (FastAPI) for cross-service work.