Stack & Responsibilities:
- AWS (CloudWatch) · Docker · Redis · Git
- Language: TypeScript/Node.js, C#/.NET, or Java — candidate-selected
- DB: candidate-selected relational or document store (PostgreSQL, MySQL, DynamoDB, MongoDB Atlas) with documented transactional guarantees
- Broker: candidate-selected (SQS, RabbitMQ, Kafka); observability: candidate-selected self-hosted (Prometheus+Grafana, OpenTelemetry)
- Own the client-server API contract: surface, versioning, error semantics, backward compatibility
- Own game economy services: server-authoritative probability, balance config tooling for Game Design
- Own IAP verification: App Store + Google Play receipts, idempotent entitlement grants, duplicate-webhook handling
- Own location proximity service: coordinate ingestion, inter-player distance, anti-spoofing
- Own observability: structured logs, latency/error metrics, on-call runbook; maintain API and architecture docs
- Use and document AI tools (Claude, ChatGPT, Gemini) so the workflow is reproducible by the team
Requirements & Nice to Have:
- Layered architecture: domain testable without framework or DB; stateless HTTP, idempotency, consistent error semantics
- API versioning with deprecation policy; configurable timeouts, retry budgets, pagination limits
- Integer/fixed-point for all currency ops — float = Non-Compliance; anti-abuse: idempotency keys, replay resistance, rate limiting
- Token-based sessions: issuance, validation, revocation, refresh; no secrets in source or commit history; documented rotation
- Single IAP interface for both stores; test mock replaceable; env via config
- Latency budgets: transaction isolation, indexing, N+1; profiling via traces, query plans, load tests
- Structured logs with correlation IDs; durable async messaging with idempotent consumers
- Nice to have: F2P live ops · anti-fraud/anti-cheat · feature flags · CI/CD + IaC · location-based features · C# reading ability · technical writing