Platform Engineering

santehq.com•New York, NY

21d•$220,000 - $280,000

About The Position

Santé is the #1 operating system for local retailers, running the entire store including point of sale, integrated payments, eCommerce, inventory management, delivery-app integrations, marketing, bookkeeping, and AI Agents. The company aims to provide technology that allows local store owners to compete with national retailers. They emphasize a fast-paced environment with daily shipping and the integration of AI-assisted engineering tools into the workflow. The Platform Engineering team is responsible for the foundational systems of Santé, including Observability, Architecture, and POS/Checkout. This team's code ensures the correctness and durability of critical operations like payment processing. They are a small, high-impact team that sets technical standards for the entire engineering organization.

Requirements

5+ years building production software, with deep experience in TypeScript/Node full-stack systems
Real depth in PostgreSQL: you read query plans for fun, know when an index helps and when it hurts, and have operated replicas under load
Distributed-systems instincts — idempotency keys, transactional boundaries, race conditions, and retry semantics are second nature
Experience owning observability for a production system (Datadog, or equivalent with OpenTelemetry)
A bias for root cause over patch, and a track record of turning one-off incidents into eliminated bug classes
Comfort operating with high ownership in a small team where your decisions become the standard

Nice To Haves

Payments experience: EMV terminals, processors (Datacap or similar), preauth/capture flows, or PCI-adjacent systems
Retail or POS domain experience, especially inventory, reconciliation, or multi-location operations
Familiarity with our stack: Next.js, tRPC, Prisma, ElastiCache, Aurora/RDS/RDS Proxy, PostgreSQL / pgvector
Experience directing AI coding agents (Claude Code, Cursor, Codex, Devin) and building the guardrails that make them safe at scale
History of establishing technical standards or architectural principles adopted across teams
Experience building LLM-powered product features (we run OCR/ CV, recommendation, and agent systems in production)

Responsibilities

Own the sale write path — the transactional core that records every sale in every store — including idempotency guarantees, race-condition elimination, split-payment correctness, and graceful degradation under database stress
Drive payments reliability end to end: card terminal integrations, processor routing, preauth recovery, retries and reconciliation when the processor or the network misbehaves
Lead our observability program on Datadog — APM, RUM, and DBM with full log/trace correlation, structured logging with PII redaction, and the monitors and DORA metrics every team runs on
Hunt performance across a busy Postgres fleet: read-replica strategy, query-plan and index work, N+1 elimination, and caching layers (ElastiCache) that hold up on the year's biggest sales days
Establish architectural patterns for multi-location retailers, offline-resilient POS, and high-write transactional workloads — and drive their adoption across teams
Multiply your output with AI agents: direct, review, and productionize agent-written code as a core part of your workflow