Solution architect

Roles and Responsibilities

• End-to-end architecture ownership — Own the architecture of the management plane, the proxy/runtime, the commerce backbone, the data platform, and the dozens of microservices behind the GraphQL federation. Design for horizontal scale, regional failover, and unpredictable upstream APIs.
• Proxy and runtime design — Steward the proxy layer that authenticates, rate-limits, meters, and forwards every API request. Tune the quota-enforcement loop and the caching topology for latency, correctness, and cost.
• Data platform architecture — Hold the line on the real-time and batch data topology (Kafka on AWS MSK, ClickHouse, Spark, Snowflake, AWS DMS for CDC, Airflow on EKS). Decide what stays real-time and what stays batch, where redundancy lives, and how the usage ledger becomes a system of record that billing, search, and customer analytics can all rely on.
• API surface and federation — Shape the GraphQL federation, subgraph boundaries, RBAC enforcement, and the relationship between the federation gateway and the microservices behind it.
• Agent-native primitives — Architect the agent-facing surface: discovery, identity, programmable commerce, governance, MCP integration. Decide what is a primitive in the platform and what is product on top.
• Reliability and security posture — Set the standards for SLOs, on-call, multi-region failover, key handling, OAuth2/JWT/mTLS, and zero-trust patterns. Lead post-incident reviews for architectural issues.
• Cost and unit economics — Own the architectural inputs to cost per call. Drive vendor consolidation across monitoring, analytics, CDP, CDN, and search, and make the platform cheaper to run as it grows.
• Technical leadership — Write the ADRs, run the design reviews, and hold the bar on whether a change belongs in the platform, in a service, or not at all. Be the senior technical voice the rest of engineering pulls on, calibrating juniors on what good architecture looks like in this codebase.

What You Bring

• 10+ years on distributed systems; multiple years owning architecture for a revenue-critical platform at scale.
• Deep hands-on AWS at scale and the data platform that runs on it — Kafka, ClickHouse, Spark, Snowflake, Airflow, CDC — and the real-time vs batch trade-offs, plus Terraform for infrastructure-as-code.
• Production fluency in at least one of Node.js/TypeScript, Go, Python, or Java — Node.js/TypeScript a plus, as the proxy and Analytics Service are Node.
• Designed and run event-driven, multi-region systems where latency, fan-out, and back-pressure matter, and where billing or rate-limiting depends on the data being right.
• Track record on commerce or metering infrastructure — usage ledgers, invoicing, payouts (Stripe / PayPal or similar) — plus strong GraphQL federation intuition.
• Security and compliance comfort: OAuth2 / JWT / mTLS, secrets management, SOC2 / ISO27001.
• Bonus: exposure to agentic systems, MCP, A2A, x402, or LLM-driven workflows — the agent-native build-out depends on it.
• You write things down, design before you build, and pick the simplest architecture that survives the next 10x. CS/engineering background or equivalent experience