Job Description
Job Description
Lead Platform Engineer For B2B Marketplace (AI-Native Workflow)
Company: BRDJ
Role: Full-Time, High-Autonomy
*Who We Are*
BRDJ is a full-stack B2B procurement marketplace structurally replacing the legacy institutional supply chain. We are not just a software layer. We operate the digital marketplace that connects nursing homes directly to manufacturer floor pricing, and we own the physical distribution network that delivers the goods. We just completed our 0-to-1 MVP build and have 25 facilities live and transacting on the platform.
*The Engineering Reality*
Our MVP Technical Lead built a hardened, enterprise-grade cloud architecture at lightning speed because he didn't write every line of code by hand. This platform was architected using heavy agentic workflows (Claude Code, Cursor, etc.).
We are not looking for a traditional "typist" developer who spends four hours writing boilerplate. We are looking for an AI-Native Platform Engineer to take over the solo keys to this machine. You are a systems thinker and a conductor. You will leverage AI agents to generate code, while your actual job is Systems Architecture, Code Review, and Deployment Safety.
The Tech Stack (What You Are Taking Over)
This is a highly modern, distributed, zero-fail environment.
* Infrastructure: 100% AWS Serverless, natively managed via SST v3 (Ion) in TypeScript.
* Backend: Node.js 20, Apollo Server 4 (GraphQL).
* Database ORM: PostgreSQL 15, Prisma ORM, ElastiCache Redis 7.1, OpenSearch 2.11.
* Transaction Engine: Asynchronous Event-Driven Outbox Pattern utilizing AWS SQS Queues and Dead-Letter Queues (DLQs) for absolute payment safety.
* Frontend: Next.js 15 (App Router), React 19, Tailwind CSS 4 in a Turborepo monorepo.
* External Integrations: Stripe Connect (routing/splits), Zebra WMS native scanner sync, Sentry (observability).
*The Day-to-Day Operation*
As the solo operator of the post-MVP ecosystem, you will not have a product manager handing you 40-page briefs. You will work directly with the COO and CEO to translate physical logistics (like Universal Parent Description taxonomies and warehouse bin locations) into digital infrastructure.
Your core workflow will look like this:
Agentic Feature Building: You will execute our final feature modules (Stripe Connect splits, native WMS-lite) by directing AI agents, reviewing the output against our strict business logic, and executing zero-downtime database migrations via Prisma.
Surgical Triage (On-Call): When a production bug hits, you don't panic. You pull the stack trace directly from Sentry, feed it into your AI IDE environment, identify the exact failing resolver, review the AI-generated patch for structural safety, and push the SST deployment in 15 minutes.
Architectural Guardrails: An AI can write a React component, but it cannot independently architect an SQS Dead-Letter Queue redrive script. You are the human filter that ensures idempotency, data integrity, and strict serverless boundaries.
The Required DNA (Non-Negotiable)
You are AI-Native: You treat AI as a co-founder. You are completely fluent in prompting, context-loading, and reviewing code from agentic tools.
You are a Database Custodian: You treat live production data as holy text. You have deep experience with Prisma, PostgreSQL, and executing zero-downtime schema migrations on live systems.
You are Cloud-Fluent: You don't need a DevOps guy. You can read, modify, and scale resources natively inside an sst.config.ts file.
You think in Systems, not just Screens: You care deeply about how order-splitting logic affects physical warehouse pick-paths and how asynchronous queues prevent double-charging.