Senior Platform Engineer

Senior Platform Engineer (contract)

We’re looking for a Senior Platform Engineer (3-6 months initial contract, open to extend) to join our distributed team at Graphy.

You’ll own the infrastructure for our new product (Agents API + SDK), building it from scratch with modern tooling and a strong focus on developer experience. This is greenfield work where you’ll make foundational decisions about how we deploy, scale and operate the platform.

Graphy currently serves ~300k users, and while your primary focus will be the new platform, you’ll also provide light support to our existing infrastructure to keep things stable during the transition. We’re a small team that ships fast, so developer experience matters as much as reliability.

What you’ll do

Main focus:

• Build a new infra layer from scratch with excellent developer experience and strong resilience.
• Design for scale. We’re exploring serverless approaches (e.g. Vercel Functions), but you’ll help us evaluate other options and their trade-offs across DX, cost, reliability and long-term portability.
• Own CI/CD end-to-end (fast deploys under 5 min, true preview branches for e2e testing, blue-green, rollbacks etc).
• Create a great local dev experience with fast feedback loops.
• Set up monitoring, alerting and observability.

Secondary focus:

• Automate SDK publishing and monorepo workflows (changelogs, versioning, internal dev tools).

• Improve platform resilience by designing fallback and failover paths (e.g. routing to AWS or Anthropic if Vercel or OpenAI has issues).

• Provide light support for our existing, legacy infra to keep things stable during the transition.

• What we look for

• You take ownership of platform and infra problems end to end.

• You care deeply about developer experience and have built infra that product engineers actually enjoy using and can maintain long term.

• You’re fluent in modern tooling (e.g. SST, Vercel, PlanetScale, Pulumi, Hono).

• You think in trade-offs, not just tools, and can explain why you’d choose one approach over another.

Nice-to-haves:

• Experience with AI/LLM infrastructure and observability (e.g. Langfuse).

• Comfortable with TypeScript / Node.js and monorepos.

• Our current stack

• For the new Agents API + SDK (graphy.dev, docs.graphy.dev), you’ll help choose the right stack. We’re currently exploring Vercel Functions and Hono, but we’re open to better ideas.

• Our legacy SaaS (graphy.app and visualize.graphy.app) runs on AWS + Vercel with GitHub Actions and Terraform, supports blue-green deploys and auto-recovery and uses PlanetScale. You may occasionally support it, but it’s not the core of the role.