Backend Engineer - Senior

You'll own the backend infrastructure layer for Bounti's AI-native platform serving frontline teams. This includes API design, data modelling, asynchronous systems, AI pipelines, integrations, and the technical decisions that shape the product as it scales.

What you'll do

• Own the backend behind the product: API design, data modelling, query performance, and the decisions that get harder to reverse the longer we leave them.
• Run everything asynchronous: queues, background jobs, scheduled work, and the retry and failure behaviour that keeps a short outage from becoming a long one.
• Own the AI pipelines: the ones running today and the next set of features we build on top of them, including versioned prompts, evaluation, cost and latency budgets, and graceful behaviour when a model provider has a bad day.
• Move data where it needs to go: reporting and analytics pipelines that give operators a trustworthy picture across dozens of locations.
• Connect Bounti to the systems customers already run: POS, scheduling, payroll, and whatever comes next.
• Make failure visible and recoverable: idempotency, reconciliation, per-customer alerting, and drift detection when a partner API misreports state.
• Own the partner relationship on the technical side: sandbox access, certification programmes, API reviews.
• Build internal tooling that empowers the CS team to diagnose failed syncs and enable integrations.

What you'll need

• Built and operated production backend systems and stayed around long enough to live with your decisions.
• Think in partial failure and eventual consistency by default. Retries, idempotency, and reconciliation are your starting assumptions, not things you add after the first incident.
• Worked against APIs you did not control, and can tell specific stories about what went wrong.
• Patient with vendors, whether that is a POS company or a model provider.
• Work with agents and stay accountable for the output. We build with Claude Code and Cursor daily.
• Pragmatic about scope: you know which abstraction is worth building now and which one is speculative.
• Experience with TypeScript, Node.js, Postgres, Prisma, Kubernetes is a plus.

Nice to have

• Run LLM features in production, including evaluation and cost control
• Built observability for systems you did not fully control
• Worked through a partner certification or app review process end to end
• Experience in hospitality, retail, or workforce tooling

What they offer

• Real ownership of a capability nobody owns today
• Work with a small, high-trust team with low ego and high standards
• Great office location in Berlin with top equipment
• 30 days holiday
• Impact from day 0